ICEIS 2024 Abstracts

Area 1 - Databases and Information Systems Integration
Full Papers Short Papers
Area 2 - Artificial Intelligence and Decision Support Systems
Full Papers Short Papers
Area 3 - Information Systems Analysis and Specification
Full Papers Short Papers
Area 4 - Software Agents and Internet Computing
Full Papers Short Papers
Area 5 - Human-Computer Interaction
Full Papers Short Papers
Area 6 - Enterprise Architecture
Full Papers Short Papers

Area 1 - Databases and Information Systems Integration

Full Papers
Paper Nr: 16
Title:

The Evolution of Original ERP Customization: A Systematic Literature Review of Technical Possibilities
Authors:

Adrian Abendroth, Benedict Bender and Norbert Gronau
Abstract: Enterprise Resource Planning (ERP) system customization is often necessary because companies have unique processes that provide their competitive advantage. Despite new technological advances such as cloud computing or model-driven development, technical ERP customization options are either outdated or ambiguously formulated in the scientific literature. Using a systematic literature review (SLR) that analyzes 137 definitions from 26 papers, the result is an analysis and aggregation of technical customization types by providing clearance and aligning with future organizational needs. The results show a shift from ERP code modification in on-premises systems to interface and integration customization in cloud ERP systems, as well as emerging technological opportunities as a way for customers and key users to perform system customization. The study contributes by providing a clear understanding of given customization types and assisting ERP users and vendors in making customization decisions.
Download
Paper Nr: 28
Title:

Data-Driven Process Analysis of Logistics Systems: Implementation Process of a Knowledge-Based Approach
Authors:

Konstantin Muehlbauer, Stephan Schnabel and Sebastian Meissner
Abstract: Due to the use of planning and control systems and the integration of sensors in the material flow, a large amount of transaction data is generated by logistics systems in daily operations. However, organizations rarely use this data for process analysis, problem identification, and process improvement. This article presents a knowledge-based, data-driven approach for transforming low-level transaction data obtained from logistics systems into valuable insights. The procedure consists of five steps aimed at deploying a decision support system designed to identify optimization opportunities within logistics systems. Based on key performance indicators and process information, a system of interdependent effects evaluates the logistics system’s performance in individual working periods. Afterward, a machine learning model classifies unfavorable working periods into predefined problem classes. As a result, specific problems can be quickly analyzed. By means of a case study, the functionality of the approach is validated. In this case study, a trained gradient-boosting classifier identifies predefined classes on previously unseen data.
Download
Paper Nr: 37
Title:

A Performance Analysis for Efficient Schema Design in Cloud-Based Distributed Data Warehouses
Authors:

Fred R. Ferreira and Robson N. Fidalgo
Abstract: Data Warehouses (DWs) have become an indispensable asset for companies to support strategic decision-making. In a world where enterprise data grows exponentially, however, new DW architectures are being investigated to overcome the deficiencies of traditional relational Database Management Systems (DBMS), driving a shift towards more modern, cloud-based DW solutions. To enhance efficiency and ease of use, the industry has seen the rise of next-generation analytics DBMSs, such as NewSQL, a hybrid storage class of solutions that support both complex analytical queries (OLAP) and transactional queries (OLTP). We under-stand that few studies explore whether the way the data is denormalized has an impact on the performance of these solutions to process OLAP queries in a distributed environment. This paper investigates the role of data modeling in the processing time and data volume of a distributed DW. The Star Schema Benchmark was used to evaluate the performance of a Star Schema and a Fully Denormalized Schema in three different market solutions: Singlestore, Amazon Redshift and MariaDB Columnstore in two different memory availability scenarios. Our results show that data denormalization is not a guarantee for improved performance, as solutions performed very differently depending on the schema. Furthermore, we also show that a hybrid-storage (HTAP) NewSQL solution can outperform an OLAP solution in terms of mean execution time.
Download
Paper Nr: 39
Title:

Quantitative Analysis of the Relationship Between Master Data Quality and Process Quality
Authors:

Simon N. Vetter, Annika Zettl, Markus M. Mützel and Omid Tafreschi
Abstract: The interplay between master data quality and process quality is well-recognized across industries, yet quantifying this relationship is complex. This paper introduces a methodology for analyzing this relationship within a business context, thereby utilizing quantitative data to enhance decision-making processes. We developed a practical approach to establish metrics for measuring master data and process quality, serving as a guideline for other businesses. Central to our methodology is the application of linear regression analysis to understand the dynamics and interplay between these two factors. To validate our approach, we implemented it in a major European-based chemical enterprise with global operations, demonstrating its effectiveness and applicability in a real-world setting.
Download
Paper Nr: 52
Title:

Text-to-SQL Meets the Real-World
Authors:

Eduardo R. Nascimento, Grettel M. García, Lucas Feijó, Wendy Z. Victorio, Yenier T. Izquierdo, Aiko R. de Oliveira, Gustavo C. Coelho, Melissa Lemos, Robinson S. Garcia, Luiz P. Leme and Marco A. Casanova
Abstract: Text-to-SQL refers to the task defined as “ given a relational database D and a natural language sentence S that describes a question on D, generate an SQL query Q over D that expresses S”. Numerous tools have addressed this task with relative success over well-known benchmarks. Recently, several LLM-based text-to-SQL tools, that is, text-to-SQL tools that explore Large Language Models (LLMs), emerged that outperformed previous approaches. When adopted for industrial-size databases, with a large number of tables, columns, and foreign keys, the performance of LLM-based text-to-SQL tools is, however, significantly less than that reported for the benchmarks. This paper then investigates how a selected set of LLM-based text-to-SQL tools perform over two challenging databases, an openly available database, Mondial, and a proprietary industrial database. The paper also proposes a new LLM-based text-to-SQL tool that combines features from tools that performed well over the Spider and BIRD benchmarks. Then, the paper describes how the selected tools and the proposed tool, running under GPT-3.5 and GPT-4, perform over the Mondial and the industrial databases over a suite of 100 carefully defined natural language questions that are closely related to those observed in practice. It concludes with a discussion of the results obtained.
Download
Paper Nr: 83
Title:

KluSIM: Speeding up K-Medoids Clustering over Dimensional Data with Metric Access Method
Authors:

Larissa R. Teixeira, Igor R. Eleutério, Mirela T. Cazzolato, Marco A. Gutierrez, Agma M. Traina and Caetano Traina-Jr.
Abstract: Clustering algorithms are powerful data mining techniques, responsible for identifying patterns and extracting information from datasets. Scalable algorithms have become crucial to enable data mining techniques on large datasets. In literature, k-medoid-based clustering algorithms stand out as one of the most used approaches. However, these methods face scalability challenges when applied to massive datasets and high dimensional vector spaces, mainly due to the high computational cost in the swap step. In this paper, we propose the KluSIM method to improve the computational efficiency of the swap step in the k-medoids clustering process. KluSIM leverages Metric Access Methods (MAMs) to prune the search space, speeding up the swap step. Additionally, KluSIM eliminates the need of maintaining a distance matrix in memory, successfully overcoming memory limitations in existing methodologies. Experiments over real and synthetic data show that KluSIM outperforms the baseline FasterPAM, with a speed up of up to 881 times, requiring up to 3,500 times fewer distance calculations, and maintaining a comparable clustering quality. KluSIM is well-suited for big data analysis, being effective and scalable for clustering large datasets.
Download
Paper Nr: 85
Title:

Analyzing Sepsis Treatment Variations in Subpopulations with Process Mining
Authors:

F. M. Rademaker, R. H. Bemthuis, J. J. Arachchige and F. A. Bukhsh
Abstract: Healthcare processes frequently deviate from established treatment protocols due to unforeseen events and the complexities of illnesses. Many healthcare procedures do not account for variations in treatment paths across different diseases and patient subpopulations. Understanding the similarities and differences in treatment paths for different patient groups can provide valuable insights and potential process enhancements for various subgroups of concern. For hospitals, understanding various patient populations, such as severe or non-severe cases, is key for enhancing care paths. In this paper, we aim to compare treatment procedures for different subpopulations of patients using process mining techniques and identify indicators to improve the care path. We utilize the process mining for healthcare (PM2 HC) methodology to identify variations in treatment paths among different patient subgroups. We conducted a case study on sepsis, a complex illness with a wealth of available data, for in-depth analysis. Our findings indicate that various subpopulations exhibit different outcomes, offering promising directions for further research.
Download
Paper Nr: 90
Title:

Similarity-Slim Extension: Reducing Financial and Computational Costs of Similarity Queries in Document Collections in NoSQL Databases
Authors:

William Z. Silva, Igor R. Eleutério, Larissa R. Teixeira, Agma M. Traina and Caetano Traina Júnior
Abstract: Several popular cloud NoSQL data stores, such as MongoDB and Firestore, organize data as document collections. However, they provide few resources for querying complex data by similarity. The comparison conditions provided to express queries over documents are based only on identity, containment, or order relationships. Thus, reading through an entire collection is often the only way to execute a similarity query. This can be both computationally and financially expensive, because data storage licenses charge for the number of document reads and writes. This paper presents Similarity-Slim, an innovative extension for NoSQL databases, designed to reduce the financial and computational costs associated with similarity queries. The extension was evaluated on the Firestore repository as a case study, considering three application scenarios: geospatial, image recommendation and medical support systems. Experiments have shown that it can reduce costs by up to 2,800 times and speed up queries by up to 85 times.
Download
Paper Nr: 91
Title:

Scoping: Towards Streamlined Entity Collections for Multi-Sourced Entity Resolution with Self-Supervised Agents
Authors:

Leonard Traeger, Andreas Behrend and George Karabatis
Abstract: Linking multiple entities to a real-world object is a time-consuming and error-prone task. Entity Resolution (ER) includes techniques for vectorizing entities (signature), grouping similar entities into partitions (blocking), and matching entity pairs based on specified similarity thresholds (filtering). This paper introduces scoping as a new and integral phase in multi-sourced ER with potentially increased heterogeneity and more unlinkable entities. Scoping reduces the space of candidate entity pairs by ranking, detecting, and removing unlinkable entities through outlier algorithms and reusable self-supervised autoencoders, leaving intact the set of true linkages. Evaluations on multi-sourced schemas show that autoencoders perform best in schemas relevant to each other, where they reduce entity collections to 77% and still contain all linkages.
Download
Paper Nr: 113
Title:

A Computer Vision-Based Method for Collecting Ground Truth for Mobile Robot Odometry
Authors:

Ricardo M. Santos, Mateus C. Silva and Ricardo R. Oliveira
Abstract: With the advancement of artificial intelligence and embedded hardware development, the utilization of various autonomous navigation methods for mobile robots has become increasingly feasible. Consequently, the need for robust validation methodologies for these locomotion methods has arisen. This paper presents a novel ground truth positioning collection method relying on computer vision. In this method, a camera is positioned overhead to detect the robot’s position through a computer vision technique. The image used to retrieve the positioning ground truth is collected synchronously with data from other sensors. By considering the camera-derived position as the ground truth, a comparative analysis can be conducted to develop, analyze, and test different robot odometry methods. In addition to proposing the ground truth collection methodology in this article, we also compare using a DNN to perform odometry using data from different sensors as input. The results demonstrate the efficacy of our ground truth collection method in assessing and comparing different odometry methods for mobile robots. This research contributes to the field of mobile robotics by offering a reliable and versatile approach to assess and compare odometry techniques, which is crucial for developing and deploying autonomous robotic systems.
Download
Paper Nr: 135
Title:

Advancing Industry 4.0: Integrating Data Governance into Asset Administration Shell for Enhanced Interoperability
Authors:

Mario Angos-Mediavilla, Michael Gorenzweig, Gerome Pahnke, André Pomp, Matthias Freund and Tobias Meisen
Abstract: The concept of Asset Administration Shell (AAS) is gaining attention in both the scientific community and manufacturing enterprises within the context of digital transformation and Industry 4.0. AAS enables the digital representation of information and services related to assets, facilitating their use and optimization in specific use cases. Standardization and the use of AAS as a vehicle for data transfer enables the collaborative exchange of information between value chain participants throughout the product life cycle. In this sense, it is essential to define and conceptualize the data governance (DG) aspects necessary to enable the use of the AAS concept in industry. Despite its significance, this topic has so far been insufficiently addressed in the scientific community. Therefore, this paper aims to identify the relevant aspects of DG needed in the AAS ecosystem, through a literature review. Based on these identified aspects, this paper addresses in detail, access control, role and rights management, and data management principles. Next, we suggest solutions for integrating these conceptual approaches into the current AAS metamodel. This approach lays the foundation for the adoption of AAS in industry, encouraging standardized data sharing practices among industry stakeholders.
Download
Paper Nr: 160
Title:

A Regression Deep Learning Approach for Fashion Compatibility
Authors:

Luís Silva, Ivan Gomes, C. M. Araújo, Tiago Cepeda, Francisco Oliveira and João Oliveira
Abstract: In the ever-evolving world of fashion, building the perfect outfit can be a challenge. We propose a fashion recommendation system, which we call Visual Search, that uses computer vision and deep learning to ensure that it has a co-ordinated set of fashion recommendations. It looks at photos of incomplete outfits, recognizes existing items, and suggests the most compatible missing piece. At the heart of our system lies a compatibility model made of a Convolutional Neural Network and bidirectional Long Short Term Memory to generate a complementary missing piece. To complete the recommendation process, we incorporated a similarity model, based on Vision Transformer. This model meticulously compares the generated image to the catalog items, selecting the one that most closely matches the generated image in terms of visual features.
Download
Paper Nr: 225
Title:

An Information System for Training Assessment in Sports Analytics
Authors:

Vanessa Meyer, Lena Wiese and Ahmed Al-Ghezi
Abstract: This paper presents an information system that analyzes and visualizes sports and human activity data. Clustering is used to divide data into groups; however, the wide variation in methods for data preprocessing and clustering makes it difficult to decide on appropriate methods. Thus, for the analysis of clustering methods, we comparatively evaluate methods for preprocessing the data in addition to the different methods for clustering. In addition, our sports analytics information system provides an approach that is able to assign athletes to a cluster based on their individual features and hence provides an individual training assessment compared to the clusters obtained on the data. The proposed visualization approach in comparison to a certain cluster offers an intuitive solution for assessing the goodness of fit.
Download
Short Papers
Paper Nr: 31
Title:

Generalizing Conditional Naive Bayes Model
Authors:

Sahar S. Yazdi, Fatma Najar and Nizar Bouguila
Abstract: Given the fact that the prevalence of big data continues to evolve, the importance of information retrieval techniques becomes increasingly crucial. Numerous models have been developed to uncover the latent structure within data, aiming to extract necessary information or categorize related patterns. However, data is not uniformly distributed, and a substantial portion often contains empty or missing values, leading to the challenge of ”data sparsity”. Traditional probabilistic models, while effective in revealing latent structures, lack mechanisms to address data sparsity. To overcome this challenge, we explored generalized forms of the Dirichlet distributions as priors to hierarchical Bayesian models namely the generalized Dirichlet distribution (LGD-CNB model) and the Beta-Liouville distribution (LBL-CNB model). Our study evaluates the performance of these models in two sets of experiments, employing Gaussian and Discrete distributions as examples of exponential family distributions. Results demonstrate that using GD distribution and BL distribution as priors enhances the model learning process and surpass the performance of the LD-CNB model in each case.
Download
Paper Nr: 42
Title:

From Data to Insights: Research Centre Performance Assessment Model (PAM)
Authors:

Oksana Tymoshchuk, Monica Silva, Nelson Zagalo and Lidia Oliveira
Abstract: This paper presents the Performance Assessment Model (PAM), designed to refine assessment practices for the impact of scientific projects and make them easier to understand with the help of information visualisation tools (InfoVis). The model incorporates three main dimensions: input, output, and impact, to capture the breadth of scientific contributions. Using PAM, a holistic analysis of project results and impacts can be conducted, integrating qualitative and quantitative data. The project team tested the model on ten research projects, which allowed for its adaptation to different project types and ensured a comprehensive assessment of tangible and intangible impact. Data organised with PAM was transferred to Power BI, a software that allows for interactive visualisation and detailed data analysis. The model’s adaptability and flexibility make it valuable for assessing how effectively scientific projects create positive, enduring impacts on society. The study results indicate that PAM provides a systematic approach to evaluating and enhancing the performance of scientific projects. It is particularly beneficial for research centre managers needing an effective tool to measure their projects’ impacts. PAM also promotes transparency and accountability in the evaluation process. Ultimately, it can ensure scientific projects are carried out effectively and efficiently, maximising societal benefits.
Download
Paper Nr: 46
Title:

Extending Semantic RML Mappings with Additional Source Formats
Authors:

Johannes Theissen-Lipp, Niklas Schäfer, Max Kocher, Philipp Hochmann, Michael Riesener and Stefan Decker
Abstract: Across many domains, the growing amount of data presents a challenge in extracting meaningful insights. A significant hurdle is the accurate interpretation and integration of data from diverse sources, often dictated by their specific applications. The RDF Mapping Language (RML), based on the W3C recommendation R2RML, can be used to transform heterogeneous data formats to RDF using defined mappings. However, existing RML implementations only support a limited set of (semi-)structured data sources such as CSV, SQL, XML, and JSON, neglecting numerous use-cases relying on other formats. This work overcomes this limitation by proposing a methodology to flexibly extend RML to support additional source formats. We systematically analyze RML and its implementations to derive a generic concept for the extension of RML. Our contributions include a general workflow for extending RML with new formats and demonstrative implementations of the RML Mapper for two examples from Building Information Modeling (BIM) and UML class diagrams. Leveraging open-source code forks and a demonstrative domain-specific language ensures easy portability to any other source format. The evaluation covers authoring of mappings, runtime performance, and practical applicability. The results affirm the effectiveness of our generic methodology for extending RML mappings to include additional source formats.
Download
Paper Nr: 56
Title:

GraphVault: A Temporal Graph Persistence Engine
Authors:

Julian Bichl, Thomas Driessen, Melanie Langermeier and Bernhard Bauer
Abstract: Graph structures have gained increasing popularity in recent years as they offer comprehensive possibilities for managing and analyzing high interconnected data. In order to facilitate the orchestration of these data, graph databases have been developed enabling graphs to be stored as central entity. However, traditional graph databases and frameworks consider graphs as a inherently valid unit without temporal reference which can limit their ability to perform advanced analysis. This paper presents GraphVault, a graph persistence engine that is capable of efficiently storing graphs and reconstructing labeled property graphs over time. We present our temporal data model, which we mapped to a key-value engine using a purpose-built record design. The performance of our implementation is then compared to that of a conventional graph database.
Download
Paper Nr: 60
Title:

MfCodeGenerator: A Code Generation Tool for NoSQL Data Access with ONM Support
Authors:

Evandro M. Kuszera, Leticia M. Peres and Marcos D. Fabro
Abstract: NoSQL databases are generally employed in scenarios that require horizontal scalability and flexibility in data schema. Applications can access the NoSQL database through native APIs or through ONMs (Object-NoSQL Mappers). The latter provides a uniform data access interface, decoupling the application from the database and reducing vendor lock-in. However, ONM code creation should be performed by developers and can be cumbersome and error prone. In this paper we propose an approach to generate ONM code based on a NoSQL schema that describes the structure of the entities and their relationships. From the NoSQL schema, our tool is used to generate code for three widely used Java-based ONMs. To evaluate the approach we perform experiments to read and write data to and from an existing MongoDB database using the generated code. Through the results obtained, it was possible to verify that the tool is capable of generating code according to the NoSQL schema and the requirements of the target ONM. This not only streamlines developer access to NoSQL data but also facilitates comparative evaluations of different ONMs utilizing the same schema.
Download
Paper Nr: 62
Title:

EmbedDB: A High-Performance Time Series Database for Embedded Systems
Authors:

Justin Schoenit, Seth Akins and Ramon Lawrence
Abstract: Efficient data processing on embedded devices may reduce network communication and improve battery usage allowing for longer sensor lifetime. Data processing is challenged by limited CPU and memory hardware. EmbedDB is a key-value data store supporting time series and relational data on memory-constrained devices. EmbedDB is competitive with SQLite on more powerful embedded hardware such as the Raspberry Pi and executes on hardware such as Arduinos that SQLite and other previous systems cannot. Experimental results evaluating EmbedDB on time series query processing show a speedup of five times compared to SQLite on a Raspberry Pi on many queries, and the ability to execute data processing on small embedded systems not well supported by existing databases.
Download
Paper Nr: 114
Title:

Evaluating UX Factors on Mobile Devices: A Feasibility Study
Authors:

Adriana L. Damian, Cinthia Carrenho, Graziela Martin, Lucas Castro, Bruna Brotto, Frederick Lucan and Raquel Pignatelli da Silva
Abstract: The acceptance of consumers regarding software products determines their success of technologies, making it a crucial topic in industrial research. In this context, the evaluation of User Experience (UX) can provide benefits in understanding for practitioners and researchers before the launch of products in the market. The literature encompasses works that focus on the assessment of UX for various software products, emphasizing the importance of clearly evaluating UX characteristics for those involved in a project. This paper presents a feasibility study with the participation of 25 practitioners engaged in the evaluation of UX for mobile devices, analyzing UX problems concerning different UX factors presented in the literature. The application of these factors was deemed easy and useful in understanding the quality of mobile devices before their market release. The study aims to contribute to practitioners and researchers involved in the assessment of UX for mobile devices, addressing different perspectives on product quality.
Download
Paper Nr: 124
Title:

Architecture for Stablecoins with Cross-Chain Interoperability
Authors:

Éric C. Machado, Juliana M. Bezerra and Celso M. Hirata
Abstract: Blockchain is the enabling technology that implements the operations of cryptocurrencies. Stablecoin is a type of cryptocurrency designed to reduce price volatility. This stability is achieved by tethering the value of the stablecoin to a reserve of assets, often in the form of a fiat currency like the US dollar. Implementing a stable-coin involves various technical challenges related to the design and architecture, which include smart contract complexity and cross-chain interoperability. This work presents an architecture for the backend of stablecoin services that address these two challenges. In the architecture, the bridge component enables seamless cross-chain interoperability, allowing to move of stablecoins from one blockchain to another without the need to be reverted to fiat currency. We developed a proof of concept, using the stablecoins deployed on both Ethereum and Polygon testnets. The proof of concept demonstrated that the architecture offers a design reference to implement other similar stablecoin systems.
Download
Paper Nr: 125
Title:

Knowledge Graph Generation from Text Using Supervised Approach Supported by a Relation Metamodel: An Application in C2 Domain
Authors:

Jones O. Avelino, Giselle F. Rosa, Gustavo R. Danon, Kelli F. Cordeiro and Maria C. Cavalcanti
Abstract: In the military domain of Command and Control (C2), doctrines contain information about fundamental concepts, rules, and guidelines for the employment of resources in operations. One alternative to speed up personnel (workforce) preparation is to structure the information of doctrines as knowledge graphs (KG). However, the scarcity of corpora and the lack of language models (LM) trained in the C2 domain, especially in Portuguese, make it challenging to structure information in this domain. This article proposes IDEA-C2, a supervised approach for KG generation supported by a metamodel that abstracts the entities and relations expressed in C2 doctrines. It includes a pre-annotation task that applies rules to the doctrines to enhance LM training. The IDEA-C2 experiments showed promising results in training NER and RE tasks, achieving over 80% precision and 98% recall, from a C2 corpus. Finally, it shows the feasibility of exploring C2 doctrinal concepts through an RDF graph, as a way of improving the preparation of military personnel and reducing the doctrinal learning curve.
Download
Paper Nr: 136
Title:

APOENA: Towards a Cloud Dimensioning Approach for Executing SQL-like Workloads Using Machine Learning and Provenance
Authors:

Raslan Ribeiro, Rafaelli Coutinho and Daniel de Oliveira
Abstract: Over the past decade, data production has accelerated at a fast pace, posing challenges in processing, querying, and analyzing huge volumes of data. Several platforms and frameworks have emerged to assist users in handling large-scale data processing through distributed and HPC environments, including clouds. Such platforms offer a plethora of cloud-based services for executing workloads efficiently in the cloud. Among these workloads are SQL-like queries, the focus of this paper. However, leveraging these platforms usually requires users to specify the type and number of virtual machines (VMs) to be deployed in the cloud. This task is not straightforward, even for expert users, as they must choose the VM type and number from several options available in a cloud provider’s catalog. Although autoscaling mechanisms can be available, non-expert users may find it challenging to configure them. To assist non-expert users in dimensioning the cloud environment for executing SQL-like workloads in such platforms, e.g., Databricks, this paper introduces a middleware named APOENA, which is designed to dimension the cloud for specific SQL-like workloads by collecting provenance data. These data are used to train Machine Learning (ML) models capable of predicting query performance for a particular combination of query characteristics and VM configuration.
Download
Paper Nr: 139
Title:

Exploring Popular Software Repositories: A Study on Sentiment Analysis and Commit Clustering
Authors:

Bianca R. Vieira and Rogério E. Garcia
Abstract: The software repositories store data and metadata about the project development, including commits, which record user modifications to projects and their metadata, such as the user responsible for the commit, date, time, and others. The programmer can register a comment to inform the modification content, its purpose, requester, motivation, and useful data. Focusing on those comments, this paper proposes using comments to group the commits and construct a sentiment analysis regarding the messages. The main purpose is to analyze those messages, both by the groups and the sentiments expressed, to understand them (what sort of sentiment they express). Opinions are central to almost all human activities and are key influences on our behaviors. Beliefs, perceptions of reality, and choices made are conditioned upon sentiments. Therefore, understanding how the developers, especially programmers, feel about a task might be useful in analyzing progress and interaction among people and artifacts (source code). In this paper, we present initial analyses of data and metadata from the twenty most popular software repositories, written in five popular programming languages. We stated five research questions and answered them, pointing out further investigations.
Download
Paper Nr: 144
Title:

Analyzing Spatial Data with Heuristics Methods and Ensemble: A Case Study of Vehicle Routing Problem
Authors:

Giovani Farias, Timotio Cubaque, Eder Gonçalves and Diana Adamatti
Abstract: The vehicle routing problem presents an intricate challenge within logistics and cargo transport. The primary objective is to determine the most efficient vehicle routes to visit a designated set of clients while minimizing overall transportation costs. The capacitated vehicle routing problem represents a specific variation of this challenge, introducing constraints such as routes commencing and concluding at the same depot, assigning each client to a single vehicle, and ensuring that the total demand for a route does not exceed the vehicle’s capacity. This paper explores the hypothesis that optimal optimization strategy is contingent on spatial data density. Thereby, we evaluate various routing strategies using heuristic methods and ensemble techniques applied to spatial data. The goal is to identify the most effective strategy tailored to a specific spatial data pattern. To accomplish this, we employ two clustering methods – K-means and DBSCAN – to group clients based on their geographical locations. Additionally, we utilize the nearest neighbor heuristic to generate initial solutions, which are subsequently refined through the implementation of the 2-Opt method. Through experiments, we demonstrate the impact of each approach on the resulting routes, taking into account the spatial data distribution.
Download
Paper Nr: 145
Title:

From Tracking Lineage to Enhancing Data Quality and Auditing: Adding Provenance Support to Data Warehouses with ProvETL
Authors:

Matheus Vieira, Thiago de Oliveira, Leandro Cicco, Daniel de Oliveira and Marcos Bedo
Abstract: Business intelligence processes running over Data Warehouses (BIDW) heavily rely on quality, structured data to support decision-making and prescriptive analytics. In this study, we discuss the coupling of provenance mechanisms into the BIDW Extract-Transform-Load (ETL) stage to provide lineage tracking and data auditing, which (i) enhances the debugging of data transformation and (ii) facilitates issuing data accountability reports and dashboards. These two features are particularly beneficial for BIDWs tailored to assist managers and counselors in Universities and other educational institutions, as systematic auditing processes and accountability delineation depend on data quality and tracking. To validate the usefulness of provenance in this domain, we introduce the ProvETL tool that extends a BIDW with provenance support, enabling the monitoring of user activities and data transformations, along with the compilation of an execution summary for each ETL task. Accordingly, ProvETL offers an additional BIDW analytical layer that allows visualizing data flows through provenance graphs. The exploration of such graphs provides details on data lineage and the execution of transformations, spanning from the insertion of input data into BIDW dimensional tables to the final BIDW fact tables. We showcased ProvETL capabilities in three real-world scenarios using a BIDW from our University: personnel admission, public information in paycheck reports, and staff dismissals. The results indicate that the solution has contributed to spotting poor-quality data in each evaluated scenario. ProvETL also promptly pinpointed the transformation summary, elapsed time, and the attending user for every data flow, keeping the provenance collection overhead within milliseconds.
Download
Paper Nr: 167
Title:

Bibliometric Insights into Web Scraping and Advanced AI-Based Models for Valuable Business Data
Authors:

Barba Giuliana, Lazoi Mariangela and Lezzi Marianna
Abstract: The integration of advanced Artificial Intelligence (AI) based models with web scraping technique opens new opportunities for businesses, streamlining the extraction of valuable insights from the huge amounts of online data. This integration is strategic in overcoming the challenges of extracting dirty data and retrieving missing information, which could otherwise compromise the reliability of business decisions. Despite the growing importance of integrating AI-based models and web scraping techniques in the business context, there exists a significant gap in understanding the specific implications. To address this gap, our study uses a systematic literature review (SLR) and bibliometric analysis to examine the implications of the combined use of advanced AI-based models and web scraping in business contexts. The study highlights four distinct clusters that suggest potential research areas in the areas of “Machine Learning (ML) for sentiment analysis”, “Artificial Intelligence and Natural Language Processing (NLP) integration”, “Data intelligence and optimization”, “NLP and Deep Learning (DL) integration”. The paper offers both theoretical and practical contributions, providing a clear overview of emerging research directions in the field of AI-based models and web scraping integration and guiding managers in adopting advanced AI-based models to enhance the value of web data obtained through scraping.
Download
Paper Nr: 176
Title:

A Hybrid Framework for Resource-Efficient Query Processing by Effective Utilization of Existing Resources
Authors:

Mayank Patel and Minal Bhise
Abstract: Scientific experiments and contemporary applications generate substantial volumes of data daily, posing a challenge for traditional database management systems (DBMS) that expend considerable time and resources on data loading. In-situ engines offer a distinct advantage by enabling immediate querying on raw data. Re-searchers have observed that resources are often underutilized during data loading. In contrast, in-situ engines spend ample time and resources in reparsing required data multiple times. Allocating query specific resources is another challenging task that must be addressed to reduce overall workload execution time and resource utilization. This research paper introduces a novel approach called the Resource Availability & Workload-aware Hybrid Framework (RAW-HF), designed to enhance the efficiency of data querying by judiciously utilizing optimal resources in systems comprising an in-situ engine and DBMS. RAW-HF incorporates modules that facilitate the optimization of resources necessary for executing a given workload, striving to maximize the utilization of available resources. The effectiveness of RAW-HF is demonstrated using the scientific dataset Sloan Digital Sky Survey (SDSS) and Linked Observation data (LOD). Comparative analysis with the state-of-the-art workload-aware partial loading technique (WA) reveals that RAW-HF excels in allocating query-specific resources and implementing resource-aware task scheduling. Results from the study indicate that RAW-HF outperforms WA, reducing workload execution time by 26%. It also reduces CPU and IO resource utilization by 26% and 25% compared to WA at a cost of 33% additional RAM.
Download
Paper Nr: 178
Title:

Blockchain Applied to Security in Industrial Internet of Things Devices
Authors:

Paulo H. Mariano, Charles B. Garrocho, Carlos F. Cavalcanti and Ricardo R. Oliveira
Abstract: The combination of blockchain and the Industrial Internet of Things brings a set of possibilities in Industry 4.0, allowing the implementation of robust and intelligent cyber-physical systems. An important issue is that the system must be secure, ensuring that data are transmitted reliably, at a time that meets temporal requirements and is immune to cyberattacks. Despite the efficiency and innovation provided by Industrial Internet of Things devices, they face significant cybersecurity challenges due to their limited capacity and exposure to risks. This article addresses aspects of data security in industrial applications in the context of Industry 4.0, understanding that the blockchain is a robust and affordable solution that offers immutability and data decentralization. Through a literature review, we examine the benefits and challenges of blockchain adoption, such as its scalability and integration with limited devices. The study points to the need for future research into the practical application of blockchain in Industrial Internet of Things environments, evaluating its effectiveness against complex cyberattacks.
Download
Paper Nr: 209
Title:

A Methodology for Constructing Patterns for the Management of Data Science Projects
Authors:

Christian Haertel, Sarah Schramm, Matthias Pohl, Sascha Bosse, Daniel Staegemann, Christian Daase and Klaus Turowski
Abstract: In the era of Big Data, the successful completion of Data Science (DS) projects is crucial. However, DS project management is quite challenging due to its interdisciplinary nature. Existing DS process models, such as CRISP-DM, have limitations, resulting in low success rates for these undertakings. To address this issue, a novel methodology for the construction of patterns in DS project management has been proposed, using the Design Science Research methodology. The design draws inspiration from existing pattern concepts to address common problems in DS project execution. The methodology is demonstrated through the creation of patterns for best practices in DS project management, synthesized from scientific literature. The goal of this approach is to provide a platform for exchanging and standardizing best practices in DS project management. While initial demonstrations show the general applicability of the methodology, further evaluations and case studies are necessary to assess its effectiveness and areas for improvement. The study identifies potential ambiguities in certain activities within the process, suggesting opportunities for refinement. Overall, this research contributes to the field of DS project management by offering a structured method to encapsulate and disseminate effective practices, supporting the successful execution of data projects in organizations.
Download
Paper Nr: 223
Title:

The Future of Oil and Gas Offloading: Leveraging Blockchain for Enhanced Transparency and Efficiency
Authors:

Paulo H. Alves, Isabella Frajhof, Élisson M. Araújo, Rafael Nasser, Gustavo Robichez, Cristiane Lodi, Carlos H. Fernandes, Rhenan Borges and Gilson Lopes
Abstract: In the dynamic and complex arena of the oil and gas sector, the management of offloading activities presents considerable challenges, particularly regarding data transparency, distribution, security, and financial transactions involving multiple parties, e.g., companies in a joint venture. The nature of multiple-party environments requires a high level of systematization, transparency, and activity orchestration to manage these challenges effectively. To address these challenges, this paper explores an innovative solution employing blockchain technology, creating efficient mechanisms to enhance transparency and the security of recorded transaction. The solution specifically focuses on the processes of oil production recording, lifting schedule management, and the intricate handling of loans and refunds. We underscore the criticality of managing loans and refunds to facilitate the lifting process, ensuring equitable oil volume distribution among consortium members. Thus, this work presents a comprehensive blockchain-based system that provides the accuracy and integrity of data, enhancing transparency and trust among consortium participants. This system seamlessly integrates all stages of offloading operations, from planning to execution, thereby revolutionizing crucial data management practices in the oil and gas sector by applying blockchain technology. Our findings suggest that implementing such technology in this context fosters a collaborative, trustworthy, secure, and efficient operational environment.
Download
Paper Nr: 226
Title:

A Unified Teaching Platform for (No)SQL Databases
Authors:

Vanessa Meyer, Lena Wiese and Ahmed Al-Ghezi
Abstract: Databases form the basic backend for information systems. This paper describes the development of a digital learning tool to promote learning of (No)SQL databases like PostgreSQL, Cassandra, Neo4J and MongoDB and the underlying data models using the React library. The learning tool will be uniformly connected to each of the mentioned databases. Thus, students can enter and execute their database queries, which are needed to solve tasks for a given example scenario, directly in our learning tool. This allows students to fully concentrate on learning the respective query languages. In this study, we present the web application’s architecture and front-end design, which will be continuously extended with additional components, such as a learning analytics dashboard. With this approach we want to contribute to the improvement of teaching methods in the field of databases and create a basis for the further development of interactive learning tools.
Download
Paper Nr: 231
Title:

Coping with Artificial Intelligence Ethical Dilemma and Ethical Position Choices?
Authors:

Sylvie Gerbaix, Sylvie Michel and Marc Bidan
Abstract: The aim of this conceptual article is to demonstrate that proposing measures, actions, and decisions to improve the ethics of Artificial Intelligence (AI) depends on the ethical theoretical position chosen. To achieve this, we proceeded in two stages. Firstly, we characterized and synthesized three different ethical issues posed by AI. Secondly, we selected two main ethical positions proposed by philosophical literature. Finally, we showed that the choice of an ethical theoretical position for each category of ethical issues of AI leads to different decisions. We demonstrated that for each category of ethical problems, the ethical decisions and their consequences differ depending on the ethical theory chosen. The value of this paper is to highlight that the literature on AI ethics often neglects the implications of choosing an ethical position. In order to attempt to solve ethical issues, it is necessary to reach agreements and have discussions that take into account the different ethical theoretical positions and their consequences in terms of decision-making.
Download
Paper Nr: 234
Title:

Empowering Multidimensional Machine Learning over Cloud- Enabled Big Data Infrastructures with ClustCube
Authors:

Alfredo Cuzzocrea, Carmine Gallo and Marco A. Mastratisi
Abstract: Multidimensional Machine Learning is emerging as one of the key features in the whole Big Data Analytics landscape. Within this broad context, the OLAP paradigm is a reference pillar, and it represents the theoretical and methodological foundation of the so-called Multidimensional Big Data Analytics trend, an emerging trend in the Big Data era. In this paper, we show how the state-of-the-art ClustCube framework, which predicates the marriage between OLAP and Clustering methodologies, can be successfully used and exploited for effectively and efficiently supporting Multidimensional Big Data Analytics in real-life big data applications and systems.
Download
Paper Nr: 246
Title:

Integrated Data Repository System: Fusion, Learning and Sharing
Authors:

Jeferson Lopes, Giancarlo Lucca, Rafael Huszcza, Amanda Mendes, Eduardo N. Borges, Pablo B. Guilherme and Leandro A. Pereira
Abstract: Currently, an enormous volume of data is being generated from diverse sources, including sensors and social media. Effectively managing this unprecedented scale of data and deriving meaningful insights from these extensive datasets present a significant challenge for computer scientists. In this context, this paper outlines the development and documentation of a project dedicated to actively contributing to these critical data-driven initiatives. The described system integrates the features of a scientific data repository with a suite of data science methods, machine learning tools, and resources for geographic data visualization. By consolidating these functionalities on a single platform, users can streamline their workflow and extract insights from data more efficiently. This integrated approach facilitates seamless transitions from data storage to model training and analysis, fostering collaboration and facilitating knowledge sharing among researchers and practitioners. In this work, we highlight the system’s key features, focusing on the datasets repository and the machine learning module as central components of our platform.
Download
Paper Nr: 248
Title:

A New Product’s Demand Forecasting Using Artificial Neural Network
Authors:

Natapat Areerakulkan, Chanicha Moryadee, Lamphai Trakoonsanti, Martusorn Khaengkhan and Natpatsaya Setthachotsombut
Abstract: This paper presents the means to improve new product (mobile phone) demand forecasting that led to total cost reduction and more efficient inventory management. The selected forecast methods, namely Holt-Winters (HW), Autoregressive Integrated Moving Average (ARIMA), Exponential Smoothing (ETS), and Artificial Neural Network (ANN), are implemented, where the most accurate method, ANN is selected to forecast demand of the new product (sixth generation mobile phone) for the following year. In addition, the comparison between the original and ANN method shows that ANN is 51.28% more accurate. After that, we develop the proposed solution plan that links improved demand forecasting to calculate the suitable inventory quantities and production rates for both finished goods and work in process. The proposed solution scenario when compared with problem scenario can reduce loss sales and inventory carrying costs by $1,400,626.80 or equivalent to 27.71%.
Download
Paper Nr: 36
Title:

The Power of Information Visualization for Understanding the Impact of Digital Media Projects
Authors:

Mónica Silva, Lersi Duran, Sofia Bermudez, Fábio Ferreira, Oksana Tymoshchuk, Lídia Oliveira and Nelson Zagalo
Abstract: This study aims to understand the most effective way to present the results and impacts of research projects in the field of Digital Media collected by the Digital Media Observatory. The focus is developing dashboards using InfoVis tools and Business Intelligence to showcase a large volume of collected data, with a team of Students@DigiMedia. Takes an exploratory approach with three main phases: researching available InfoVis tools, creating sample dashboards using InfoVis tools, and implementing project dashboards using Power BI. The team of students has developed dashboards that provide a clear, structured view of the project, aggregating the following information: title, logo, objectives, keywords, funding, human resources, partners, methodological procedures, scientific and technological products, publications, dissemination, recognition, and SDGs. These dashboards provide interactive reports and visualisations to help researchers analyse, and communicate project results. This study can help to improve the overall data presentation experience, simplifying the analysis and knowledge-sharing process within the digital media research community.
Download
Paper Nr: 45
Title:

Heterogeneous Data Integration: A Literature Scope Review
Authors:

Silvia L. Borowicc and Solange N. Alves-Souza
Abstract: Data have been collected by communities for analysis, visualization, predictions and other activities to support data-driven decision. Obtaining value from data assets directly depends on the data integration task. However, Big Data poses new challenges to integration due to data heterogeneity. It is essential to understand the main problems and to know technologies and techniques that have been employed to improve the ability to obtain value by heterogeneous data integration. This paper presents a literature scope review that highlights the main techniques applied to heterogeneous data integration. The literature reviewed presents solutions mostly focusing on a specific purpose or part of the integration process instead of a clear understanding of how the techniques can be used in a complete integration process. Therefore, this work shows a whole picture of a data integration process organizing the techniques according to their functionalities and presents a workflow with tasks associated to techniques and resources, focusing on semantic mediation, such as mapping and matching tasks. Ontologies and semantic web technologies are promising to address data heterogeneity and have been used in the semantic enrichment of data and semantic mediation between data sources and global model. However, some aspects remain to be further investigated, such as ontology and terminology construction, data processing scalability and semantic mediation, especially for mapping definition.
Download
Paper Nr: 50
Title:

What to Do when Privacy Issues Screw It Up: Ingestion Refactoring in a Big-Data Pipeline
Authors:

Gian P. Jesi, Nicola Spazzoli, Andrea Odorizzi and Gianluca Mazzini
Abstract: Privacy is an increasingly important concern especially in the European Union (EU). With the growing use of technology individuals’ personal information is being collected and processed by companies and organizations on a massive scale. In order to be compliant with Privacy regulations and General Data Protection Regulation (GDPR) in particular, we could no longer use a software tool, LOGGIT, for several reasons. This tool was a cornerstone in one of our Big-data pipeline ingestion. We did our best to comply with this requirement as soon as possible. In this work, we discuss how we refactor our pipeline architecture several times in order to find a balance between our requirements in terms of reliability and the regulations of the GDPR.
Download
Paper Nr: 53
Title:

An Alternative Way to Analyze and Predict Consonant Clusters Productions in Brazilian Portuguese Phonological Assessments
Authors:

João B. Marques, João D. Lima, Márcia Keske-Soares and Fabrício A. Rubin
Abstract: To conduct phonological assessments in children, it is necessary to have a set of words that contains a representative sample of adult vocabulary. One of the obstacles to obtain a minimal set is the need to include words with various consonant clusters so that such complex phonetic structures can be validated. In the current literature, there is only one way to determine whether a child is capable of producing a consonant cluster: through the application of a phonological assessment, which contains several words with diverse phonetic structures to be evaluated. In this context, logical inferences are one of the fundamental pillars in any learning area, as they establish logical connections between information to form knowledge about a specific subject. This work proposes an alternative way to indirectly assess a child’s ability to produce consonant clusters, based on their ability to articulate similar clusters. The proposed algorithm is fed with the consonant clusters produced and not produced by the child during the assessment. The goal is to discern which other clusters the child is capable or incapable of producing, using the separation of consonant clusters into simpler phonetic structures. The method was validated with a database containing over 1200 phonological assessments conducted in school-age children, native speakers of Brazilian Portuguese. The accuracy of our approach was 97% with 12% false positives and 8% false negatives, indicating that the method is interesting and significantly faithful to real-world results but still leaves room for future improvements. Nevertheless, it is believed that it can be used to reduce the number of words needed in a phonological assessment, through indirect evaluation of specific phonetic structures.
Download
Paper Nr: 98
Title:

A Comprehensive Blockchain-Based Architecture for Healthcare Systems
Authors:

José R. Melo, Inaldo C. Costa, Juliana M. Bezerra and Celso M. Hirata
Abstract: Blockchain technology has emerged as a versatile solution with wide-ranging applications across various industries, including healthcare. The increasing number of breaches in medical records in health systems highlights the imperative for innovative solutions. This paper delves into the potential of blockchain to improve information management in healthcare systems, considering data privacy, cybersecurity, and reliability concerns. We propose a blockchain-based architecture that takes into account key entities of healthcare systems, such as patients, physicians, diagnostic centers, and pharmacies, and facilitates their transactions through the use of blockchain technology. Through comprehensive sequence diagrams, we illustrate the orchestrated interactions among selected entities. The paper presents a proof of concept implementation, providing details on application development, smart contract specifications facilitating seamless information sharing, and the tests conducted. The implementation of the blockchain-based architecture and sequence diagrams was successfully tested. We conclude that the proposed architecture enables the improvement of the data privacy of entities, the cybersecurity of data sharing among diverse entities, and the reliability of transactions within healthcare systems.
Download
Paper Nr: 108
Title:

A Framework for Organisational Readiness Assessment in Digital Business Ecosystems Engagement
Authors:

Ruimian Li and Kecheng Liu
Abstract: Worldwide competition is forcing companies to collaborate in digital business ecosystems (DBEs) in order to leverage resources and survive in the global market. However, the engagement of companies in DBEs is confronted by a number of practical issues. This research has as its objective the discovery of the critical factors and the framework that enable organisations to assess their readiness in engaging in DBEs to cooperate with their peers. To accomplish the objective, this research has explained the related concepts and theories and developed a research framework grounded on a theoretical and literature review background. The assessment results help identify specific key weakness for the companies to improve themselves to implement DBE engagement in the future.
Download
Paper Nr: 174
Title:

CWM Extensions for Knowledge and Metadata Integration for Complex Data Warehouse and Big Data
Authors:

Ralaivao J. Christian, Razafindraibe Fabrice, Raherinirina Angelo and Rakotonirainy Hasina
Abstract: This document constitutes a continuation of the work carried out in the field of complex data warehouses (DW) relating to the management and formalization of knowledge and metadata. It proposes a methodological approach to integrate two concepts, knowledge and metadata, within the framework of a complex DW architecture. The objective of the work considers the use of the knowledge representation technique by description logic and the extension of the Common Warehouse Metamodel (CWM) specifications. Several essential aspects of this work are expected, including the representation of knowledge in description logics and the declination of this knowledge into coherent UML diagrams while respecting or extending CWM specifications and using XML as a pivot, in particular OWL DL. Furthermore, the coupling between UML Ontology Profile (UOP) and the Ontology Definition Metamodel (ODM), for semantic modeling, integration of ontologies or enrichment of metadata, will be operationalized by transformation of models or by mapping or both simultaneously. As a result, a new extension of CWM metamodel will be developed. This will have performance consequences for a complex DW. The field of application is vast but will be adapted to systems with heterogeneous, complex and unstructured content and requiring a large (re)use of knowledge such as medical data warehouses.
Download
Paper Nr: 238
Title:

Implementation of Composable Enterprise in an Evolutionary Way Through Holistic Business-IT Delivery of Business Initiatives: Real Industry Use Case
Authors:

Ivka Ivas
Abstract: Service composability, introduced by service-oriented architecture (SOA), is a design principle that encourages the design of reusable services that themselves also consist of reusable services. In domain driven design (DDD), which inspired microservice architectures, the scope of composable service design is interpreted as a software solution domain, while the problem domain lies in the detached business world. This results in IT solutions that are often redundant at the enterprise level or tend to be composable only within a specific enterprise IT ecosystem as a result of the design without understanding the business domain or how the new solution fits into the overall delivery and enterprise architecture. On the other hand, it is not uncommon for company´s "business", motivated by revenue increase, to push frequent deliveries of business changes, putting pressure on company´s IT to implement quick fix solutions that only solve immediate business problems. All this leads to inconsistent and redundant software systems that increase the complexity of the organization and result in higher maintenance costs and less flexibility in implementing future changes. As a solution, this paper proposes Composable Enterprise, a business-IT approach for architecting the enterprise that introduces Business Composability and a holistic understanding of the enterprise. Business Composability is a business-IT-aligned service abstraction that starts with the notion of first applying service composability to business assets (business capabilities) to achieve the scale and pace required to realize business changes. The purpose of this paper is to provide a methodology for implementing Composable Enterprise in large, complex organisations, not as a massive, enterprise-wide rationalization and consolidation initiative, but in an evolutionary way through the joint and holistic business-IT delivery of business initiatives. The application of the proposed methodology is illustrated using a real-industry use case.
Download
Paper Nr: 259
Title:

Datasets on Mobile App Metadata and Interface Components to Support Data-Driven App Design
Authors:

Jonathan C. Kuspil, João S. Ribeiro, Gislaine L. Leal, Guilherme C. Guerino and Renato Balancieri
Abstract: The global mobile device market currently encompasses 6.5 billion users. Therefore, standing out in the competitive scenario of application stores such as the Google Play Store (GPlay) requires, among several factors, great concern with the User Interface (UI) of the apps. Several datasets explore UI characteristics or the metadata present in GPlay, which developers and users write. However, few studies relate these data, limiting themselves to specific aspects. This paper presents the construction, structure, and characteristics of two Android app datasets: the Automated Insights Dataset (AID) and the User Interface Depth Dataset (UID). AID compiles 48 different metadata from the 200 most downloaded free apps in each GPlay category, totaling 6400 apps, while UID goes deeper into identifying 7540 components and capturing 1948 screenshots of 400 high-quality apps from AID. Our work highlights clear selection criteria and a comprehensive set of data, allowing metadata to be related to UI characteristics, serving as a basis for developing predictive models and understanding the current complex scenario of mobile apps, helping researchers, designers, and developers.
Download

Area 2 - Artificial Intelligence and Decision Support Systems

Full Papers
Paper Nr: 15
Title:

Graph Convolutional Networks for Image Classification: Comparing Approaches for Building Graphs from Images
Authors:

Júlia P. Rodrigues and Joel L. Carbonera
Abstract: Graph Neural Networks (GNNs) is an approach that allows applying deep learning techniques to non-euclidean data such as graphs and manifolds. Over the past few years, graph convolutional networks (GCNs), a specific kind of GNN, have been applied to image classification problems. In order to apply this approach to image classification tasks, images should be represented as graphs. This process usually involves over-segmenting images in non-regular regions called superpixels. Thus, superpixels are mapped to graph nodes that are characterized by features representing the superpixel information and are connected to other nodes. However, there are many ways of transforming images into graphs. This paper focuses on the use of graph convolutional networks in image classification problems for images over-segmented into superpixels. We systematically evaluate the impact of different approaches for representing images as graphs in the performance achieved by a GCN model. Namely, we analyze the degree of segmentation, the set of features chosen to represent each su-perpixel as a node, and the method for building the edges between nodes. We concluded that the performance is positively impacted when increasing the number of nodes, considering rich sets of features, and considering only connections between similar regions in the resulting graph.
Download
Paper Nr: 59
Title:

The Traveling Tournament Problem: Rows-First versus Columns-First
Authors:

Kristian Verduin, Ruben Horn, Okke van Eck, Reitze Jansen, Thomas Weise and Daan van den Berg
Abstract: At the time of writing, there is no known deterministic time algorithm to uniformly sample initial valid solutions for the traveling tournament problem, severely impeding any evolutionary approach that would need a random initial population. Repeatedly random sampling initial solutions until we find a valid one is apparently the best we can do, but even this rather crude method still requires exponential time. It does make a difference however, if one chooses to generate initial schedules column-by-column or row-by-row.
Download
Paper Nr: 79
Title:

Improving Explainability of the Attention Branch Network with CAM Fostering Techniques in the Context of Histological Images
Authors:

Pedro L. Miguel, Alessandra Lumini, Giuliano C. Medalha, Guilherme F. Roberto, Guilherme B. Rozendo, Adriano M. Cansian, Thaína A. Tosta, Marcelo Z. do Nascimento and Leandro A. Neves
Abstract: Convolutional neural networks have presented significant results in histological image classification. Despite their high accuracy, their limited interpretability hinders widespread adoption. Therefore, this work proposes an improvement to the attention branch network (ABN) in order to improve its explanatory power through the gradient-weighted class activation map technique. The proposed model creates attention maps and applies the CAM fostering strategy to them, making the network focus on the most important areas of the image. Two experiments were performed to compare the proposed model with the ABN approach, considering five datasets of histological images. The evaluation process was defined via quantitative metrics such as coherency, complexity, confidence drop, and the harmonic average of those metrics (ADCC). Among the results, the proposed model through the ResNet-50 was able to provide an improvement of 4.16% in the average ADCC metric and 3.88% in the coherence metric when compared to the respective ABN model. Considering the DesneNet-201 network as the explored backbone, the proposed model achieved an improvement of 14.87% in the average ADCC metric and 9.77% in the coherence metric compared to the corresponding ABN model. The contributions of this work are important to make the results via computer-aided diagnosis more comprehensible for clinical practice.
Download
Paper Nr: 110
Title:

An Evaluation of Pre-Trained Models for Feature Extraction in Image Classification
Authors:

Erick S. Puls, Matheus V. Todescato and Joel L. Carbonera
Abstract: In recent years, we have witnessed a considerable increase in performance in image classification tasks. This performance improvement is mainly due to the adoption of deep learning techniques. Generally, deep learning techniques demand a large set of annotated data, making it challenging when applied to small datasets. Transfer learning strategies have become a promising alternative to overcome these issues in this scenario. This work compares the performance of different pre-trained neural networks for feature extraction in image classification tasks. We evaluated 16 different pre-trained models in four image datasets. Our results demonstrate that the best general performance along the datasets was achieved by CLIP-ViT-B and ViT-H-14, where the CLIP-ResNet50 model had similar performance but with less variability. Therefore, our study provides evidence supporting the choice of models for feature extraction in image classification tasks.
Download
Paper Nr: 117
Title:

Out of Sesame Street: A Study of Portuguese Legal Named Entity Recognition Through In-Context Learning
Authors:

Rafael O. Nunes, Andre S. Spritzer, Carla S. Freitas and Dennis G. Balreira
Abstract: This paper explores the application of the In-Context Learning (ICL) paradigm for Named Entity Recognition (NER) within the Portuguese language legal domain. Identifying named entities in legal documents is complex due to the intricate nature of legal language and the specificity of legal terms. This task is important for a range of applications, from legal information retrieval to automated summarization and analysis. However, the manual annotation of these entities is costly due to the specialized knowledge required from legal experts and the large volume of documents. Recent advancements in Large Language Models (LLM) have led to studies exploring the use of ICL to improve the performance of Generative Language Models (GLMs). In this work, we used Sabiá, a Portuguese language LLM, to extract named entities within the legal domain. Our goal was to evaluate the consistency of these extractions and derive insights from the results. Our methodology involved using a legal-domain NER corpus as input and selecting specific samples for a prompting task. We then instructed the GLM to catalog its own NER corpus, which we compared with the original test examples. Our study examined various aspects, including context examples, selection strategies, heuristic methodologies, post-processing techniques, and quantitative and qualitative analyses across specific domain classes. Our results indicate promising directions for future research and applications in specialized domains.
Download
Paper Nr: 121
Title:

Optimizing Planning Strategies: A Machine Learning Forecasting Model for Energy Aggregators and Hydropower Producers
Authors:

Sarah Di Grande, Mariaelena Berlotti, Salvatore Cavalieri and Roberto Gueli
Abstract: The global push for higher renewable energy production is driven by concerns about climate change, pollution, and diminishing fossil fuel reserves. Governments, businesses, and communities worldwide prioritize cleaner energy sources like solar, wind, and hydroelectric, over traditional fuels. Technological advancements enhancing efficiency and cost-effectiveness have made renewables more competitive, catalyzing their growing dominance in the energy market. In this context, renewable energy forecasting models are fundamental for both operators of the energy market called energy aggregators, and prosumers for different reasons like planning, decision-making, energy sales optimization, and investment evaluation. Therefore, the present work aimed to develop a machine learning model designed for multi-step hydropower forecasting of plants integrated into Water Distribution Systems (WDSs). The Alcantara 1 Hydroelectric Plant, situated in Italy, was utilized as the case study. This plant generates electricity from the water flow utilized for municipal water supply, which is then sold to the medium voltage network, resulting in substantial remuneration. This innovative approach utilizes previously unused architectures like TCN and N-Beats, to provide multi-step hydropower forecasting for WDS-integrated plants, a special category of systems for which models have not yet been developed. Results indicate TCN as the most accurate model for addressing the proposed task.
Download
Paper Nr: 191
Title:

SpectraNet: A Neural Network for Soybean Contents Prediction
Authors:

Henry J. Kobs, Henrique L. Krever, Denilson S. Ebling and Celio Trois
Abstract: Soybeans are integral to global agriculture and food production, playing a vital role in human and animal nutrition. Accurate assessment of moisture, oil, and protein contents in soybeans is crucial for various applications, including human nutrition, animal feed, and food manufacturing. This paper introduces SpectraNet, a Neural Network architecture designed for predicting soybean contents using Near-infrared Spectroscopy (NIRS) data. NIRS technology provides a cost-effective and non-destructive means of analyzing agricultural samples. Spec-traNet leverages a 1D convolutional Neural Network and multiple prediction heads, demonstrating its efficacy in handling non-linearities present in spectral data. The architecture’s flexibility and adaptability contribute to accurate predictions, automatic feature extraction, and suitability for varying conditions. Comparative analysis with traditional Partial Least Squares Regression (PLSR) models reveals the superior performance of SpectraNet in predicting protein, moisture, and oil contents in soybeans. The presented methodology involves comprehensive data collection, laboratory analysis, and model training, showcasing the potential of SpectraNet for real-world applications in agriculture. The results highlight the efficiency and precision of SpectraNet, offering a valuable tool for advancing agricultural practices and ensuring soybean quality.
Download
Paper Nr: 202
Title:

The Power of Gyroscope Data: Advancing Human Movement Analysis for Walking and Running Activities
Authors:

Patrick N. Alvim, Jonathan C. F. da Silva, Vicente P. Amorim, Pedro O. Lazaroni, Mateus C. Silva and Ricardo R. Oliveira
Abstract: The ability to faithfully reproduce the real world in the virtual environment is crucial to provide immersive and accurate experiences, opening doors to significant innovations in areas such as simulations, training, and data analysis. In such a way that actions in the virtual environment can be applied, which would be challenging in the real world due to issues of danger, complexity, or feasibility, enabling the study of these actions without compromising these principles. Additionally, it is possible to capture real-world data and analyze it in the virtual environment, faithfully reproducing real actions in the virtual realm to study their implications. However, the volatility of real-world data and the accurate capture and interpretation of such data pose significant challenges in this field. Thus, we present a system for real data capture aiming to virtually reproduce and classify walking and running activities. By using gyroscope data to capture the rotation of axes in the lower human limbs movement, it becomes possible to precisely replicate the motion of these body parts in the virtual environment, enabling detailed analyses of the biomechanics of such activities. In our observations, in contrast to quaternion data that may have different scales and applications depending on the technology used to create the virtual environment, gyroscope data has universal values that can be employed in various contexts. Our results demonstrate that, by using specific devices such as sensors instead of generic devices like smartwatches, we can capture more accurate and localized data. This allows for a granular and precise analysis of movement in each limb, in addition to its reproduction. This system can serve as a starting point for the development of more precise and optimized devices for different types of human data capture and analysis. Furthermore, it proposes creating a communication interface between the real and virtual worlds, aiming to accurately reproduce an environment in the other. This facilitates data for in-depth studies on the biomechanics of movement in areas such as sports and orthopedics.
Download
Paper Nr: 221
Title:

Use of Custom Videogame Dataset and YOLO Model for Accurate Handgun Detection in Real-Time Video Security Applications
Authors:

Diego Bazan, Raul Casanova and Willy Ugarte
Abstract: Research has shown the ineffectiveness of video surveillance operators in detecting crimes through security cameras, which is a challenge due to their physical limitations. On the other hand, it was shown that computer vision, although promising, faces difficulties in real-time crime detection due to the large amount of data needed to build reliable models. This study presents three key innovations: a gun dataset extracted from the Grand Theft Auto V game, a computer vision model trained on this data, and a video surveillance application that employs the model for automatic gun crime detection. The main challenge was to collect images representing various scenarios and angles to reinforce the computer vision model. The video editor of the Grand Theft Auto V game was used to obtain the necessary images. These images were used to train the model, which was implemented in a desktop application. The results were very promising, as the model demonstrated high accuracy in detecting gun crime in real time. The video surveillance application based on this model was able to automatically identify and alert about criminal situations on security cameras.
Download
Paper Nr: 228
Title:

A Comparative Analysis of EfficientNet Architectures for Identifying Anomalies in Endoscopic Images
Authors:

Alexandre P. Pessoa, Darlan P. Quintanilha, João D. Sousa de Almeida, Geraldo Braz Junior, Anselmo C. de Paiva and António Cunha
Abstract: The gastrointestinal tract is part of the digestive system, fundamental to digestion. Digestive problems can be symptoms of chronic illnesses like cancer and should be treated seriously. Endoscopic exams in the tract make detecting these diseases in their initial stages possible, enabling an effective treatment. Modern endoscopy has evolved into the Wireless Capsule Endoscopy procedure, where patients ingest a capsule with a camera. This type of exam usually exports videos up to 8 hours in length. Support systems for specialists to detect and diagnose pathologies in this type of exam are desired. This work uses a rarely used dataset, the ERS dataset, containing 121.399 labelled images, to evaluate three models from the EfficientNet family of architectures for the binary classification of Endoscopic images. The models were evaluated in a 5-fold cross-validation process. In the experiments, the best results were achieved by EfficientNetB0, achieving average accuracy and F1-Score of, respectively, 77.29% and 84.67%.
Download
Paper Nr: 230
Title:

Teaching Practice Using ChatGPT in Higher Education
Authors:

Edna D. Canedo, Angelica S. Calazans, Geovana S. Silva, Eloisa S. Masson and Fábio L. Mendonça
Abstract: ChatGPT is an Artificial Intelligence (AI) chatbot platform developed by OpenAI. Several studies have highlighted the advantages and disadvantages of integrating ChatGPT into teaching methodologies and knowledge generation within higher education. We conducted a survey involving 86 professors within the Computer Science field in Brazil. Our findings indicate that professors are utilizing ChatGPT for content generation and the creation of teaching materials, including practical exercises, slides, assignments, and tests. Moreover, they view ChatGPT as a potential facilitator of learning by fostering interaction between students and professors. In the realm of knowledge production, professors are leveraging ChatGPT for tasks such as aiding in the composition of research papers or articles and generating automatic summaries. However, as per the professors’ perceptions, a notable limitation of ChatGPT is its inability to provide bibliographic references for the content it delivers. Most professors believe that ChatGPT can be used as a support tool in higher education to generate knowledge. However, it is essential to address the challenges associated with the lack of bibliographic references in the content provided by ChatGPT.
Download
Paper Nr: 241
Title:

An Approach for Privacy-Preserving Mobile Malware Detection Through Federated Machine Learning
Authors:

Giovanni Ciaramella, Fabio Martinelli, Francesco Mercaldo, Christian Peluso and Antonella Santone
Abstract: Considering the diffusion of smart devices and IoT devices, mobile malware detection represents a task of fundamental importance, considering the inefficacy of signature-based antimalware free and commercial software, which can detect a threat only if its signature is present in the antimalware repository. In the last few years, many methods have been proposed by academia to identify so-called zero-day malware through machine learning: these techniques typically extract a series of features from the mobile device to send to a server where the detection model is located. Typically, these features include network traces or installed applications, among other information that may compromise user privacy. In this context, Federated learning is emerging with privacy advantages because the raw data never leaves the local device. In this paper, we propose a method to integrate federated machine learning in malware detection.Malicious software typically aims to extract sensitive and private data, and mobile devices emerge as particularly enticing targets from the perspective of attackers. In the experimental analysis, comprising a pool of 10 clients from which 7 are uniformly sampled at each round, we demonstrate the efficacy of the proposed method by achieving an accuracy of 0.940.
Download
Short Papers
Paper Nr: 26
Title:

Making Hard(er) Benchmark Functions: Genetic Programming
Authors:

Dante Niewenhuis, Abdellah Salhi and Daan van den Berg
Abstract: TreeEvolver, a genetic programming algorithm, is used to make continuous mathematical functions that give rise to 3D landscapes. These are then empirically tested for hardness by a simple evolutionary algorithm, after which TreeEvolver mutates the functions in an effort to increase the hardness of the corresponding landscapes. Results are wildly diverse, but show that traditional continuous benchmark functions such as Branin, Easom and Martin-Gaddy might not be hard at all, and much harder objective landscapes exist.
Download
Paper Nr: 27
Title:

Information Extraction in the Legal Domain: Traditional Supervised Learning vs. ChatGPT
Authors:

Gustavo C. Coelho, Alimed Celecia, Jefferson de Sousa, Melissa Lemos, Maria J. Lima, Ana Mangeth, Isabella Frajhof and Marco Casanova
Abstract: Information Extraction is an important task in the legal domain. While the presence of structured and machine-processable data is scarce, unstructured data in the form of legal documents, such as legal opinions, is largely available. If properly processed, such documents can provide valuable information about past lawsuits, allowing better assessment by legal professionals and supporting data-driven applications. This paper addresses information extraction in the Brazilian legal domain by extracting structured features from legal opinions related to consumer complaints. To address this task, the paper explores two different approaches. The first is based on traditional supervised learning methods to extract information from legal opinions by essentially treating the extraction of categorical features as text classification and the extraction of numerical features as named entity recognition. The second approach takes advantage of the recent popularization of Large Language Models (LLMs) to extract categorical and numerical features using ChatGPT and prompt engineering techniques. The paper demonstrates that while both approaches reach similar overall performances in terms of traditional evaluation metrics, ChatGPT substantially reduces the complexity and time required along the process.
Download
Paper Nr: 38
Title:

A Knowledge Base of Argumentation Schemes for Multi-Agent Systems
Authors:

Carlos A. Ferreira, Débora C. Engelmann, Rafael H. Bordini, Joel L. Carbonera and Alison R. Panisson
Abstract: Argumentation constitutes one of the most significant components of human intelligence. Consequently, argumentation has played a significant role in the community of Artificial Intelligence, in which many researchers study ways to replicate this intelligent behaviour in intelligent agents. In this paper, we describe a knowledge base of argumentation schemes modelled to enable intelligent agents’ general (and domain-specific) argumentative capability. To that purpose, we developed a knowledge base that not only enables agents to reason and communicate with other software agents using a computation model of arguments, but also with humans, using a natural language representation of arguments which results from natural language templates modeled alongside their respective argumentation scheme. To illustrate our approach, we present a scenario in the legal domain where an agent employs argumentation schemes to reason about a crime, deciding whether the defendant intentionally committed the crime or not, a decision that could significantly impact the severity of the sentence handed down by a legal authority. Once a conclusion is reached, the agent provides a natural language explanation of its reasoning.
Download
Paper Nr: 61
Title:

Occupational Accidents Prediction in Brazilian States: A Machine Learning Based Approach
Authors:

J. M. Toledo and Thiago M. Moura
Abstract: Occupational accident is an unexpected event connected to work that may result in injury and/or death of workers. Thus, the possibility of predicting the occurrence of occupational accidents can assist the government in labor policy-making, protecting the lives and health of workers. In this work, we propose the use of machine learning models to predict the occurrence of occupational accidents in each Brazillian state. We use multiple datasets concerning socio-economic, employment, and demographic data as sources to obtain an integrated table utilized to train regression models (linear regression, support vector regressor, and LightGBM) and make predictions. We verify that the developed models show high predictive performance and explainability, with the R-squared metric reaching 0.90.
Download
Paper Nr: 75
Title:

A Decision Support System Based on a Mixed-Integer Linear Programming Model for Location of Routers in Open-Pit Mines
Authors:

Matheus F. Mandarino, Tatianna P. Beneteli and Luciano P. Cota
Abstract: In open-pit mines, it is very important to ensure network coverage for equipment in operation, which is located in large areas. It should be noted that some of this equipment, such as trucks and drills, is autonomous; therefore, access to the network is essential. This work presents a mathematical model for solving it based on the p-median problem. The objective is to determine the location of the routers, minimizing the number of routers and the sum of the distances between the operating points and the installed routers. We use real data from the Fábrica Nova mine in Brazil to validate the mathematical model. The scenarios represent the mining planning for 2023, 2024, and 2025. The results showed that the proposed model found the optimal router location in a few seconds, providing more efficient coverage for mining equipment using fewer routers.
Download
Paper Nr: 82
Title:

A Visual Analysis Approach to Static Postural Control Acquired by a Force Plate
Authors:

Thales B. Uê, Danilo M. Eler and Iracimara A. Messias
Abstract: Force plates are biomechanical equipment responsible for providing data to understand the mechanics of human movement. However, mathematical software used to process data are a barrier to researchers without much experience and prior knowledge on areas from Exact Sciences and Information Technology. This paper aims to implement a visual approach to analyze human static postural control obtained from a force plate as a means of helping researchers interpret its data. By measuring ground reaction forces and their respective torque moments, the displacements of the Center of Pressure in its medial‑lateral and anterior-posterior directions are calculated to observe and evaluate the postural balance’s behavior. Data processing and visualization were implemented using Python programming language. Scatter plots, heat maps, violin plots, and box plots were used as graphic representations for data collected before and after muscular intervention in older adults with sarcopenia. Applying the developed approach makes it possible to visually observe each of the Center of Pressure’s oscillation values measured for data collection and how they relate. This fact differs from statistical information, which summarizes the sample’s data in a quantified value. Therefore, data visualization is essential to complement the statistical data and provide another view to force plate data.
Download
Paper Nr: 97
Title:

Exploratory Data Analysis in Cloud Computing Environments for Server Consolidation via Fuzzy Classification Models
Authors:

Rafael R. Bastos, Vagner A. Seibert, Giovani P. Maia, Bruno M. P. de Moura, Giancarlo Lucca, Adenauer C. Yamin and Renata S. Reiser
Abstract: The present work addresses the challenges of flexible resource management in Cloud Computing, emphasizing the critical need for efficient resource utilization. Precisely, we tackle the problem of dynamic server consolidation, supported by the capacity of Fuzzy Logic to deal with uncertainties and imprecisions inherent in cloud environments. In the preprocessing step, we employ a feature selection strategy to perform attribute selection and, better understand the problem. Data classification was performed by fuzzy rule learning approaches. Comparative evaluations of algorithm classification highlight the remarkable accuracy of FURIA, with IVTURS as a close alternative. While FURIA generates 41 rules, indicating a comprehensive model, IVTURS produces only six, introducing an abstract level to model uncertainties as interval-valued fuzzy membership degrees. The study underscores the relevance of parameter adaptation in mapping feature selection and membership functions to achieve optimal performance for flexible algorithms in the Cloud Computing environment. Our results underlie the structure of a fuzzy system adapted to CloudSim, integrating energy optimization and Service Level Agreements assurance through different server consolidation strategies. This research contributes valuable perspectives to decision-making processes in the Cloud Computing environment.
Download
Paper Nr: 103
Title:

Artificial Intelligence in Sustainable Smart Cities: A Systematic Study on Applications, Benefits, Challenges, and Solutions
Authors:

Simone D. Santos, Jéssyka F. Vilela, Thiago H. Carvalho, Thiago C. Rocha, Thales B. Candido, Vinícius S. Bezerra and Daniel J. Silva
Abstract: In an era marked by rapid urban growth and environmental challenges, the advent of “smart cities” holds promise for a sustainable future. Central to the operational efficiency of these cities is the role of Artificial Intelligence (AI). This Systematic Literature Review addresses the critical question: How can AI be used in sustainable smart cities? Using Kitchenham’s guidelines, the review followed a three-step Planning, Conducting, and Reporting process. Through a comprehensive search in the databases ACM, IEEEXplore, Scopus, Science Direct, and Emerald, a total of 46 high-quality papers were identified. These papers were scrutinized to understand the AI services utilized in smart cities, the benefits, the challenges of implementation, and potential solutions to these challenges. Findings reveal that AI’s impact is multi-dimensional, affecting transportation, energy management, and citizen engagement, among other areas. However, several challenges remain, considering ethics and data management. This review serves as an exhaustive guide for researchers and policymakers interested in leveraging AI for sustainable urban development.
Download
Paper Nr: 106
Title:

X-GAN: Generative Adversarial Networks Training Guided with Explainable Artificial Intelligence
Authors:

Guilherme B. Rozendo, Alessandra Lumini, Guilherme F. Roberto, Thaína A. Tosta, Marcelo Zanchetta do Nascimento and Leandro A. Neves
Abstract: Generative Adversarial Networks (GANs) create artificial images through adversary training between a generator (G) and a discriminator (D) network. This training is based on game theory and aims to reach an equilibrium between the networks. However, this equilibrium is hardly achieved, and D tends to be more powerful. This problem occurs because G is trained based on only a single value representing D’s prediction, and only D has access to the image features. To address this issue, we introduce a new approach using Explainable Artificial Intelligence (XAI) methods to guide the G training. Our strategy identifies critical image features learned by D and transfers this knowledge to G. We have modified the loss function to propagate a matrix of XAI explanations instead of only a single error value. We show through quantitative analysis that our approach can enrich the training and promote improved quality and more variability in the artificial images. For instance, it was possible to obtain an increase of up to 37.8% in the quality of the artificial images from the MNIST dataset, with up to 4.94% more variability when compared to traditional methods.
Download
Paper Nr: 118
Title:

Explainable Machine Learning for Alarm Prediction
Authors:

Kalleb M. Abreu, Julio S. Reis, André D. Santos and Giorgio Zucchi
Abstract: This paper evaluates machine learning models for the prediction of alarms using geographical clustering, exploring data from an Italian company. The models encompass a spectrum of algorithms, including Naive Bayes (NB), XGBoost (XGB), and Multilayer Perceptron (MLP), coupled with encoding techniques, and clustering methodologies, namely COOP (Coopservice) and KPP (K-Means++). The XGB models emerge as the most effective, yielding the highest AP (Average Precision) values across models based on MLP and NB. Hyperparameter tuning for XGB models reveals default values perform well. Our model explainability analyses reveal the significant impact of geographical location (cluster) and the time interval when the predictions are made. Challenges arise in handling dataset imbalances, impacting minority alarm class predictions. the insights gained from this study lay the groundwork for future investigations in the field of geographical alarm prediction. The identified challenges, such as imbalanced datasets, offer opportunities for refining methodologies. As we move forward, a deeper exploration of one-class algorithms holds promise for addressing these challenges and enhancing the robustness of predictive models in similar contexts.
Download
Paper Nr: 132
Title:

multiBERT: A Classifier for Sponsored Social Media Content
Authors:

Kshitij S. Malvankar, Enda Fallon, Paul Connolly and Kieran Flanagan
Abstract: Social media’s rise has given birth to a new class of celebrities called influencers. People who have amassed a following on social media sites like Twitter, YouTube, and Instagram are known as influencers. These people have the ability to sway the beliefs and purchase choices of those who follow them. Consequently, companies have looked to collaborate with influencers in order to market their goods and services. But as sponsored content has grown in popularity, it has becoming harder to tell if a piece is an independent opinion of an influencer or was sponsored by a company. This study investigates the use of machine learning models to categorise influencer tweets as either sponsored or unsponsored. By utilising transformer language models, like BERT, we are able to discover relationships and patterns between a brand and an influencer. Machine learning algorithms may assist in determining if a tweet or Instagram post is a sponsored post or not by examining the context and content of influencer tweets and their Instagram post captions. To evaluate data from Instagram and Twitter together, this work presents a novel method that compares the models while accounting for performance criteria including accuracy, precision, recall, and F1 score.
Download
Paper Nr: 148
Title:

A Machine Learning Workflow to Address Credit Default Prediction
Authors:

Rambod Rahmani, Marco Parola and Mario A. Cimino
Abstract: Due to the recent increase in interest in Financial Technology (FinTech), applications like credit default prediction (CDP) are gaining significant industrial and academic attention. In this regard, CDP plays a crucial role in assessing the creditworthiness of individuals and businesses, enabling lenders to make informed deci-sions regarding loan approvals and risk management. In this paper, we propose a workflow-based approach to improve CDP, which refers to the task of assessing the probability that a borrower will default on his or her credit obligations. The workflow consists of multiple steps, each designed to leverage the strengths of different techniques featured in machine learning pipelines and, thus best solve the CDP task. We employ a comprehensive and systematic approach starting with data preprocessing using Weight of Evidence encoding, a technique that ensures in a single-shot data scaling by removing outliers, handling missing values, and making data uniform for models working with different data types. Next, we train several families of learning models, introducing ensemble techniques to build more robust models and hyperparameter optimization via multi-objective genetic algorithms to consider both predictive accuracy and financial aspects. Our research aims at contributing to the FinTech industry in providing a tool to move toward more accurate and reliable credit risk assessment, benefiting both lenders and borrowers.
Download
Paper Nr: 152
Title:

Identification and Attribution of Access Roles Using Hierarchical Team Permission Analysis
Authors:

Iryna Didinova and Karel Macek
Abstract: This paper addresses the challenges of Role-Based Access Control (RBAC) in large organizations, with a focus on the efficient attribution of access roles. It critiques traditional role-mining algorithms and the use of Machine Learning (ML) models, which serve as benchmarks due to their lack of practical interpretability and potential security vulnerabilities. The novel contribution of this work is the introduction of the Hierarchical Team Permission Analysis (HTPA), a methodology grounded in organizational hierarchy. HTPA is shown to outperform the benchmark approaches by creating meaningful, interpretable roles that enhance both the security and efficiency of access control systems in large enterprises. The paper advocates for the potential integration of HTPA with ML models to further optimize role attribution and suggests avenues for future research in this evolving field.
Download
Paper Nr: 159
Title:

An Open-Source Approach for Digital Prostate Cancer Histopathology: Bringing AI into Practice
Authors:

Markus Bauer, Lennart Schneider, Marit Bernhardt, Christoph Augenstein, Glen Kristiansen and Bogdan Franczyk
Abstract: The histopathological analysis of prostate tissue is challenging due to the required expertise and the inherently high number of samples. This accounts especially for prostate cancer (PCa) assessment (tumour grading), as parameters like the Gleason score have high prognostic relevance, but suffer from significant interobserver variability, mainly due to individual grading practice and experience. AI-based solutions could assist pathological workflows, but their integration into clinical practice is still hampered, as they’re optimised based on general AI-metrics, rather than clinical relevance and applicability. Moreover, commercial solutions often provide similar performance than academic approaches, are expensive, and lack flexibility to adapt to new use cases. We investigate the requirements to provide a flexible AI-based histopathological tissue analysis tool, that makes the expertise of experienced pathologists accessible to every hospital in a user-friendly, open-source solution. The proposed software allows for slide inspection, tumour localisation and tissue metric extraction, while adapting to different use cases using a Python-enabled architecture. We demonstrate the value of our tool in an in-depth evaluation of transurethral hyperplastic resection tissue (TURP)-chip analysis and PCa grading using a set of extensively annotated prostate cancer patient cases. Our solution can support pathologists in challenging cases, fasten routine tasks and creates space for detail analysis.
Download
Paper Nr: 164
Title:

Swarm Intelligence Path-Planning Pipeline and Algorithms for UAVs: Simulation, Analysis and Recommendation
Authors:

Wyatt Harris, Sean Tseng, Tabatha Viso, Max Weissman and Chun-Kit Ngan
Abstract: This research work aims to support domain experts in the selection of proper path planning algorithms for UAVs to solve a domain business problem (i.e., the last mile delivery of goods). In-depth analysis, insight, and recommendations of three promising approaches, including reinforcement learning-based, bio-inspired-based, and physics-based are used to address the multi-agent UAV path planning problem. Specifically, the contributions of this work are fourfold: First, we develop a unified pipeline to implement each approach to conduct this analysis. Second, we build a 2D UAV path planning environment to simulate each approach. Third, using this 2D environment, we run the 450 simulations in three different group sizes of swarm UAV agents (i.e., 3, 5, and 10) within three environments of varying complexity (i.e., Easy, Intermediate, and Hard). We aggregate the simulation data and compare their performance in terms of success rate, run-time, and path length while using the classical A* Search as a baseline. Finally, based upon the performance of each approach and our analytical investigations, we provide informed recommendations for the optimal use case of each UAV path planning approach. The recommendations are presented using parameters for environmental complexity and urgency of goods delivery.
Download
Paper Nr: 172
Title:

Toward Air Quality Fuzzy Classification
Authors:

Vagner A. Seibert, Rafael Bastos, Giovani Maia, Giancarlo Lucca, Helida Santos, Adenauer Yamin and Renata R. Reiser
Abstract: This work considers different fuzzy classifier models to evaluate the air quality of indoor spaces, providing flexible systems related to the imprecision of metrics and parameters since the modeling process. Air Quality is a relevant topic concerning modern society, and the research on air quality evaluation provides important alternatives for improving global environmental governance. In this paper, we discuss the performances of the five fuzzy classifiers named CHI, FURIA, WF-C, FARC-HD, and SLAVE, applied in the data classification from an open dataset from Germany. Thus, this domain knowledge enables us to model the inherent uncertainties of attributes’ problems related to Air Quality and Air Quality Index. The results showed that fuzzy approaches offer a valid alternative for determining and correctly classifying indoor air quality with satisfying accuracy, adding flexible modeling in the air quality analysis.
Download
Paper Nr: 177
Title:

Knowledge-Based Systems for Strengthening African Health Systems
Authors:

Wendgounda F. Ouedraogo and Andreas Nürnberger
Abstract: The set of difficulties characterizing health systems in Sub-Saharan Africa (SSA) are caused by many factors, among others, the very limited number of specialists, the predominance of paramedical personnel and micro-health centers, the poor distribution of large medical centers, the almost absence of continuing training and missing means of maintaining and updating knowledge. These difficulties, however, find a response in the development of targeted intelligent solutions such as expert systems capable of providing continuous assistance to professionals and structures of health systems in African countries and transforming them into effective organizations. Unfortunately, the design of such systems still face many challenges, e.g., they have to promote circulation and simplified access to current targeted information (e.g. current knowledge) that healthcare professionals need in their daily lives and they have to take into account local needs to be better adapted and useful for the heterogenous group of users. This article provides a short overview of the current urgent needs of the health systems in SSA, motivates requirements on targeted digital support technology and discusses first prototypical solutions to motivate possible research directions.
Download
Paper Nr: 186
Title:

Explainable Business Intelligence for Video Analytics in Retail
Authors:

Christian Daase, Christian Haertel and Klaus Turowski
Abstract: This paper explores research questions and perspectives for the next stage of societal development, often referred to as Society 5.0, and the field of modern retail. Artificial intelligence (AI) is seen as a key component that provides retailers with the means to optimize their store layouts, advertising campaigns, and overall business strategy. The need to make AI-based decisions comprehensible and tangible to ensure acceptance by the respective target groups has been emphasized with the concept of explainable AI in various research works. Based on observations from the AI domain and the business world, the need to integrate explainability into commercial AI-driven operations is addressed and the concept of explainable business intelligence (XBI) is proposed. A set of potential research questions for video analytics in retail in the age of Society 5.0, as one of the most promising use cases in this regard, is derived from the literature and the proposal of XBI in terms of the outlined opportunities and challenges is explained, critically discussed, and visualized.
Download
Paper Nr: 188
Title:

A Fuzzy-Genetic Multi-Objective Optimization Method Applied to Deployment of Routers in Agricultural Crop Areas
Authors:

P. G. Coelho, J. M. Amaral, T. M. Carvalho, R. A. Gomes, I. S. Cardoso and T. N. Souza
Abstract: High technology is increasingly applied to improving crop fields and coined an area as Precision Agriculture. The main focus of this work is to increase production by performing data acquisition from an agricultural crop area, monitoring sensor devices to measure temperature, humidity, etc. This allows the administrator of the field to make good decisions related to the land management. This paper proposes a hybrid fuzzy-genetic multi-objective intelligent method to place routers in an agricultural crop area so that to cover the sensor monitoring devices spread over it. The method combines a genetic algorithm with a fuzzy aggregation technique to evaluate multiples objectives, in order to determine an adequate location of the routers considering the designerś preferences. Case studies are presented and show the proposal results.
Download
Paper Nr: 192
Title:

My Database User Is a Large Language Model
Authors:

Eduardo R. Nascimento, Yenier T. Izquierdo, Grettel M. García, Gustavo C. Coelho, Lucas Feijó, Melissa Lemos, Luiz P. Leme and Marco A. Casanova
Abstract: The leaderboards of familiar benchmarks indicate that the best text-to-SQL tools are based on Large Language Models (LLMs). However, when applied to real-world databases, the performance of LLM-based text-to-SQL tools is significantly less than that reported for these benchmarks. A closer analysis reveals that one of the problems lies in that the relational schema is an inappropriate specification of the database from the point of view of the LLM. In other words, the target user of the database specification is the LLM rather than a database programmer. This paper then argues that the text-to-SQL task can be significantly facilitated by providing a database specification based on the use of LLM-friendly views that are close to the language of the users’ questions and that eliminate frequently used joins, and LLM-friendly data descriptions of the database values. The paper first introduces a proof-of-concept implementation of three sets of LLM-friendly views over a relational database, whose design is inspired by a proprietary relational database, and a set of 100 Natural Language (NL) questions that mimic users’ questions. The paper then tests a text-to-SQL prompt strategy implemented with LangChain, using GPT-3.5 and GPT-4, over the sets of LLM-friendly views and data samples, as the LLM-friendly data descriptions. The results suggest that the specification of LLM-friendly views and the use of data samples, albeit not too difficult to implement over a real-world relational database, are sufficient to improve the accuracy of the prompt strategy considerably. The paper concludes by discussing the results obtained and suggesting further approaches to simplify the text-to-SQL task.
Download
Paper Nr: 199
Title:

Embedding a Data-Driven Decision-Making Work Culture in a Social Housing Environment
Authors:

Srinidhi Karthikeyan, Takao Maruyama and Sankar Sivarajah
Abstract: This paper explores the issue of delayed rent payments in social housing in the United Kingdom and its impact on tenants and housing providers. Our approach is to use machine learning algorithms to analyse payment patterns and identify tenants who may be at risk of falling behind on rent payments. By doing this, we aim to equip housing providers with the necessary tools to intervene early and maintain consistent tenancies. We have conducted research using machine learning models such as decision trees and random forests to address this issue. The paper emphasises the potential benefits of Explainable AI, which can help build trust in data-driven decision-making and AI among employees unfamiliar with AI and machine learning.
Download
Paper Nr: 208
Title:

Bridging Human and AI Decision-Making with LLMs: The RAGADA Approach
Authors:

Tapio Pitkäranta and Leena Pitkäranta
Abstract: The Retrieval Augmented Generation Algorithmic Decision Alignment (RAGADA) architecture is an advancement in AI-augmented decision-making for corporate environments. This paper discusses RAGADA’s innovative architecture that merges RAG and Multi-Agent System (MAS) with sophisticated business algorithms and dynamic interfaces, enhancing natural language interaction between AI systems and users. This fusion extends AI’s reach, facilitating adaptable decision-making tools for leaders, in line with evolving business strategies and ethical standards. Experimental validation of RAGADA within the banking sector, involving diverse stakeholder groups ranging from customers to business and ethical managers, confirms its effectiveness. The system adeptly translates natural language inquiries into actionable insights, thereby improving the user experience and decision-making transparency. This validation underscores RAGADA’s potential to transform stakeholder engagement and demonstrates a leap in utilizing AI for strategic and ethical business management.
Download
Paper Nr: 210
Title:

Using Soft Computing and Computer Vision to Create and Control an Integrated Autonomous Robotic Manipulator Process
Authors:

João T. Rodrigues, Samuel D. Anjos, Mateus C. Silva, Ricardo M. Santos and Ricardo R. Oliveira
Abstract: The development and control of an integrated autonomous robotic manipulation process requires a focus on the convergence of technologies such as soft computing, object detection, robotic arm engineering, and direct and inverse kinematics. For instance, the inverse kinematics issue can be targeted using soft computing instead of challenging mathematical applications. This paper explores using soft computing systems, an algorithm that produces approximate solutions to complex problems and phenomena. Thus, the use of soft computing proved valid, given the accuracy and speed of the claw. The soft computing technology is based on an evolutionary algorithm that allows us to create several points on a cartesian plane and mix them to implement inverse kinematics. Our results showed that using soft computing, which is different from the traditional way, leads to solid and functional results. The implementation involves integrating Arduino, Raspberry Pi 4.0, a PWM model PCA9685, a camera, and six servo motors to create a robotic arm. The system employs video streaming to transmit data to a local network, where the Raspberry Pi processes RGB to HSV images for object identification. The present work in this paper has experiments for the accuracy and speed of the claw to take a determinate object from the center and the side of the checkered board.
Download
Paper Nr: 214
Title:

Building Damage Segmentation After Natural Disasters in Satellite Imagery with Mathematical Morphology and Convolutional Neural Networks
Authors:

Antônio R. Neto and Daniel O. Dantas
Abstract: In this study, our main motivation was to develop and optimize an image segmentation model capable of accurately assessing damage caused by natural disasters, a critical challenge today where the frequency and intensity of these events are increasing. In order to predict damage categories, including no damage, minor damage, and major damage, we compared several models and approaches. we explored and compared several models, focusing on the Unet architecture employing BDANet and other architectures such as ResNet18, VGG16, and ResNet50. Layers with mathematical morphology operations were applied as a filtering strategy. The results indicated that the Unet model with the BDANet backbone had the best performance, with an F1-score of 0.761, which increased to 0.799 after applying mathematical morphology operations.
Download
Paper Nr: 217
Title:

On the Current State of Generative Artificial Intelligence: A Conceptual Model of Potentials and Challenges
Authors:

Christian Daase, Christian Haertel, Abdulrahman Nahhas, Alexander Zeier, Achim Ramesohl and Klaus Turowski
Abstract: Generative artificial intelligence (GenAI) is one of the most promising recent advances in digital technology. However, research often focuses on specific application scenarios, case studies and experiments. Overarching and comprehensive studies that consider potentials and challenges for the entire field of GenAI across domains are rather scarce. In this paper, the four domains of text, audio, image and code generation are examined by means of a systematic literature review. Opportunities for industry and society are discussed, with the aim of providing a conceptual model that enables a quick assessment of the current state-of-the-art and identifies applications for GenAI that are either not yet sufficiently researched and therefore invite further exploratory investigations, or are well researched and therefore represent recognized yet less experimental fields.
Download
Paper Nr: 227
Title:

Diffusion Model for Generating Synthetic Contrast Enhanced CT from Non-Enhanced Heart Axial CT Images
Authors:

Victor S. Ferreira, Anselmo Cardoso de Paiva, Aristofanes C. Silva, João D. Sousa de Almeida, Geraldo Braz Junior and Francesco Renna
Abstract: This work proposes the use of a deep learning-based adversarial diffusion model to address the translation of contrast-enhanced from non-contrast-enhanced computed tomography (CT) images of the heart. The study overcomes challenges in medical image translation by combining concepts from generative adversarial networks (GANs) and diffusion models. Results were evaluated using the Peak signal to noise ratio (PSNR) and structural index similarity (SSIM) to demonstrate the model’s effectiveness in generating contrast images while preserving quality and visual similarity. Despite successes, Root Mean Square Error (RMSE) analysis indicates persistent challenges, highlighting the need for continuous improvements. The intersection of GANs and diffusion models promises future advancements, significantly contributing to clinical practice. The table compares CyTran, CycleGAN, and Pix2Pix networks with the proposed model, indicating directions for improvement.
Download
Paper Nr: 239
Title:

Hydrocyclone Operational Condition Detection: Conceptual Prototype with Edge AI
Authors:

Tomás C. Silva, Ricardo R. Oliveira and Emerson Klippel
Abstract: Hydrocyclones, vital in mineral processing plants, classify materials by size and density. Operational issues, like roping, can cause inefficiencies and financial losses. This paper explores computer vision techniques for the assessment of hydrocyclone underflow operational status. Testing revealed robust performance for both a Resnet-18 and a MobileViT-V2 model. An edge device was implemented for real-time inferences on a conceptual prototype that simulates underflow scenarios. The CNN models demonstrate high precision and recall, with an F1 Score over 92% for roping detection on the edge device. The research contributes to efficient hydrocyclone monitoring, addressing challenges in remote mining locations. The findings offer potential for further optimization and industrial implementation, enhancing processing plant reliability and mitigating financial risks associated with operational irregularities.
Download
Paper Nr: 242
Title:

Studying Trustworthiness of Neural-Symbolic Models for Enterprise Model Classification via Post-Hoc Explanation
Authors:

Alexander Smirnov, Anton Agafonov and Nikolay Shilov
Abstract: Neural network-based enterprise modelling support is becoming popular. However, in practical enterprise modelling scenarios, the quantity of accessible data proves inadequate for efficient training of deep neural networks. A strategy to solve this problem can involve integrating symbolic knowledge to neural networks. In previous publications, it was shown that this strategy is useful, but the trust issue was not considered. The paper is aimed to analyse if the trained neural-symbolic models just “learn” the samples better or rely on the meaningful indicators for enterprise model classification. The post-hoc explanation (specifically, the concept extraction) has been used as the studying technique. The experimental results showed that embedding symbolic knowledge does not only improve the learning capabilities but also increases the trustworthiness of the trained machine learning models for enterprise model classification.
Download
Paper Nr: 255
Title:

Generation of Breaking News Contents Using Large Language Models and Search Engine Optimization
Authors:

João Pereira, Wenderson Wanzeller and António M. Rosado da Cruz
Abstract: With easy access to the internet, anyone can search for the latest news. However, the news found on the internet, especially on social media, are often of dubious origin. This article explores how new technologies can help journalists in their day-to-day work. We therefore sought to create a platform for generating hot news content using Artificial Intelligence, namely Large Language Models (LLMs), combined with Search Engine Optimization (SEO). We investigate how LLMs impact content production, analyzing their ability to create compelling and accurate narratives in real time. Additionally, we examine how SEO integration can optimize the visibility and relevance of this content in search engines. This work highlights the importance of strategically combining these technologies to improve efficiency in disseminating news, adapting to the dynamism of online information and the demands of a constantly evolving audience.
Download
Paper Nr: 262
Title:

A Multi-Feature Semantic Fusion and Bipartite Graph-Based Risk Identification Approach for Project Participation
Authors:

Yue Wang, Yujie Hu, Wenjing Chang and Jianjun Yu
Abstract: In the complex landscape of project management, ensuring the authenticity of participant involvement is paramount to achieving fairness, enforceability, and desired outcomes. Addressing the challenges posed by the heterogeneous nature of graphs, the underutilization of rich attribute information, and the scarcity of anomaly labels, we propose a Project Participation Authenticity Risk Identification Graph Neural Network (PARI-GNN), a novel architecture leveraging graph-based anomaly detection techniques to assess authenticity risks in project participation. PARI-GNN include a novel framework for risk identification using heterogeneous graphs. This method transforms heterogeneous graphs into bipartite graphs and combines multi-feature semantic fusion techniques with bipartite graph structures, providing a robust solution for identifying inauthentic participation. We evaluate our proposed model using real-world data. The experimental outcomes affirm the superior performance of PARI-GNN in accurately discerning authenticity risks, demonstrating the efficacy and competitive advantage of the proposed framework over a variety of state-of-the-art methodologies.
Download
Paper Nr: 69
Title:

Towards an Algorithm-Based Automatic Differentiation of Liability Cases by Analyzing Complaint Texts
Authors:

Insa Lemke and Nadine Schlüter
Abstract: Effective complaints management is important to maintain customer loyalty and offers the opportunity to feed knowledge back into product development and production. However, with products and supply chains becoming increasingly complex, the picture is often unclear when it comes to handling complaints. This applies in particular to the handling of legal liability issues. The challenge arises from the correct classification of the various legal bases in connection with the receipt of a customer complaint. To this end, we introduce the concept of an algorithm that uses automatic text recognition to analyze the text of a complaint and determine whether a liability case may exist under German law. This paper presents the different development steps and phase components of the algorithm as well as the current implementation status.
Download
Paper Nr: 72
Title:

An Integrated Decision Support System for Intra-Logistics Management with Peripheral Storage and Centralized Distribution
Authors:

Giulia Dotti, Manuel Iori, Anand Subramanian and Marco Taccini
Abstract: Intra-logistics optimization plays a crucial role in ensuring efficiency and reducing non-value added activities, especially in scenarios with a central shipping point and multiple peripheral warehouses. The goal of this study is to create an automated and optimized Decision Support System (DSS) using an integer linear programming (ILP) model. The DSS optimizes the order management process by determining optimal load configurations from peripheral warehouses onto transport vehicles. The resulting transportation plan, generated through this approach, aims to meet customer demands while minimizing overall costs. Computational tests, conducted on a real-world case study, validated the efficiency of the proposed system.
Download
Paper Nr: 104
Title:

Advanced AI-Based Solutions for Visual Inspection: A Systematic Literature Review
Authors:

Angelo Corallo, Vito Del Vecchio and Alberto Di Prizio
Abstract: Artificial Intelligence (AI)-based solutions, including Machine Learning (ML) and Deep Learning (DL), are ever more implemented in industry for assisting advanced Visual Inspection (VI) systems. They support companies in a more effective identification of product defects, enhancing the performance of humans and avoiding the risks of product incompliance. However, companies often struggle in considering the most appropriate AI-based solutions for VI and for a specific manufacturing domain. Also, an extensive literature study focused on this topic seems to lack. On the basis of a Systematic Literature Review, this paper aims to map the main advanced AI-based VI system solutions (including methods, technologies, techniques, algorithms) thus helping companies in considering the most appropriate solutions for their needs.
Download
Paper Nr: 105
Title:

Exploring Implementation Parameters of Gen AI in Companies
Authors:

Maarten Voorneveld
Abstract: Our work focusses on investigating Gen AI implementation, as the field is developing at such a rapid pace, up to date research on business implementations and outcomes is limited. We systematically evaluate AI applications, analysing challenges/opportunities. We consider adoption beyond pilot projects via a structured approach covering factors such as technological, organizational, and environmental. Our case studies show relevance of data quality, infrastructure, and organizational culture. The paper explores how company leaders can support to create employee trust and deliver on an AI strategy. Companies face competition, customer needs and regulation that shape their technology roadmaps. These complexities are exacerbated by training data problems, internal communications, context challenges and ethics. This research finds that challenges & strategies for responsible Generative AI deployment advocate a holistic and adaptive approach. Which companies need to tailor each application, to achieve desired outcome.
Download
Paper Nr: 111
Title:

Recommendation Systems: A Deep Learning Oriented Perspective
Authors:

Igor L. Lampa, Vitoria Z. Gomes and Geraldo D. Zafalon
Abstract: The massive use of the digital platforms has provided an exponential increase at the amount of data consumed and daily generated. Thus, there is a data overload which directly affects the consume experience of digital products, whether at find a news, consume an e-commerce product or to choose a movie in a streaming platform. In this context, emerge the recommendation systems, which have the finality of provide an efficient way to comprehend the user predilections and to recommend direct items. Thus, this work brings the classical concepts and techniques already used, as well as analyzes their use along with deep learning, which through evaluated results has a grater capability to obtain implicit relationships between users and items, providing recommendations with better quality and accuracy. Furthermore, considering the review of the literature and analysis provided, an architectural model for recommendation system based on deep learning is proposed, which is defined as a hybrid system.
Download
Paper Nr: 131
Title:

Optimizing Natural Language Processing Applications for Sentiment Analysis
Authors:

Anderson C. Lopes, Vitoria Z. Gomes and Geraldo D. Zafalon
Abstract: Recent technological advances have stimulated the exponential growth of social network data, driving an increase in research into sentiment analysis. Thus, studies exploring the intersection of Natural Language Processing and social network analysis are playing an important role, specially those one focused on heuristic approaches and the integration of algorithms with machine learning. This work centers on the application of sentiment analysis techniques, employing algorithms such as Logistic Regression and Support Vector Machines. The analyses were performed on datasets comprising 5,000 and 10,000 tweets, and our findings reveal the efficient performance of Logistic Regression in comparison with other approach. Logistc Regression improved the performed in almost all measures, with emphasis to accuracy, recall and F1-Score.
Download
Paper Nr: 162
Title:

Deepbrokenhighways: Road Damage Recognition System Using Convolutional Neural Networks
Authors:

Sebastian Peralta-Ireijo, Bill Chavez-Arias and Willy Ugarte
Abstract: Road damage, such as potholes and cracks, represent a constant nuisance to drivers as they could potentially cause accidents and damages. Current pothole detection in Peru, is mostly manually operated and hardly ever use image processing technology. To combat this we propose a mobile application capable of real-time road damage detection and spatial mapping across a city. Three models are going to be trained and evaluated (Yolov5, Yolov8 and MobileNet v2) on a novel dataset which contains images from Lima, Peru. Meanwhile, the viability of crack detection through bounding box method will be put to the test, each model will be trained once with cracks annotations and without. The YOLOv5 model was the one with the best results, as it showed the best mAP50 across all of out experiments. It got 99.0% and 98.3% mAP50 with the dataset without crack and with crack annotations, correspondingly.
Download
Paper Nr: 168
Title:

Towards Collective Superintelligence: Amplifying Group IQ Using Conversational Swarms
Authors:

Louis Rosenberg, Gregg Willcox, Hans Schumann and Ganesh Mani
Abstract: Swarm Intelligence (SI) is a natural phenomenon that enables biological groups to amplify their combined intellect by forming real-time systems. Artificial Swarm Intelligence (or Swarm AI) is a technology that enables networked human groups to amplify their combined intelligence by forming similar systems. In the past, swarm-based methods were constrained to narrowly defined tasks like probabilistic forecasting and multiple-choice decision making. A new technology called Conversational Swarm Intelligence (CSI) was developed in 2023 that amplifies the decision-making accuracy of networked human groups through natural conversational deliberations mediated by artificial agents. The current study evaluated the ability of real-time groups using a CSI platform to take a common IQ test known as Raven’s Advanced Progressive Matrices (RAPM). First, a baseline group of participants took the Raven’s IQ test by traditional survey. This group averaged 45.7 correct. Then, groups of approximately 35 individuals answered IQ test questions together using a CSI platform called Thinkscape. These groups averaged 80.5% correct. This puts the CSI groups in the 97th percentile of IQ test-takers and corresponds to an effective IQ increase of 28 points (p<0.001). This is an encouraging result and suggests that CSI is a powerful method for enabling conversational collective intelligence in large, networked groups. In addition, because CSI deliberations are scalable across groups of potentially any size, these methods may provide a pathway to building a Collective Superintelligence.
Download
Paper Nr: 171
Title:

Enhancing Constructivist Learning: The Role of Generative AI in Personalised Learning Experiences
Authors:

Hua Guo, Weiqian Yi and Kecheng Liu
Abstract: This paper explores the transformative role of generative AI in enhancing constructivist learning, where students actively construct knowledge through meaningful experiences. By investigating the synergies between generative AI and constructivist learning, the study uncovers how AI fosters personalized educational experiences. The research underscores the profound influence of generative AI on constructivist learning, empowering students to become active, motivated, and lifelong learners by tailoring their education, fostering creativity and collaboration, and upholding ethical principles. The study advocates for the responsible and purposeful integration of generative AI, which would revolutionize education and prepare students for future challenges.
Download
Paper Nr: 215
Title:

Glaucoma Detection Using Transfer Learning with the Faster R-CNN Model and a ResNet-50-FPN Backbone
Authors:

Noirane Getirana de Sá, Daniel O. Dantas and Gilton J. Ferreira da Silva
Abstract: Early detection of glaucoma has the potential to prevent vision loss. The application of artificial intelligence can enhance the cost-effectiveness of glaucoma detection by reducing the need for manual intervention. Glaucoma is the second leading cause of blindness and, due to its asymptomatic nature until advanced stages, diagnosis is often delayed. Having a general understanding of the disease’s pathophysiology, diagnosis, and treatment can assist primary care physicians in referring high-risk patients for comprehensive ophthalmo-logic examinations and actively participating in the care of individuals affected by this condition. This article describes a method for glaucoma detection with the Faster R-CNN model and a ResNet-50-FPN backbone. Our experiments demonstrated greater accuracy compared to models such as, AlexNet, VGG-11, VGG-16, VGG-19, GoogleNet-V1, ResNet-18, ResNet-50, ResNet-101 and ResNet-152.
Download
Paper Nr: 250
Title:

Exploring Applicability of LLM-Powered Autonomous Agents to Solve Real-life Problems: Microsoft Entra ID Administration Agent (MEAN)
Authors:

Roberto Rodriguez and Nestori Syynimaa
Abstract: Microsoft Entra ID is Microsoft’s identity and access management solution used by many public and private sector organisations globally. In March 2023, Microsoft retired two PowerShell modules which have enabled automation of administrative tasks, such as user management. The replacement module is based on Microsoft Graph API, and its effective usage would require administrators to learn software development skills. In this paper, we will report the results of work-in-progress research on exploring the applicability of LLM-powered autonomous agents to solve real-life problems. We describe the design and proof-of-concept implementation of MEAN, an agent that performs Entra ID administrative tasks using Microsoft Graph API based on natural language prompts. The results show that LLM-powered autonomous agents can perform at least simple Entra ID administrative tasks. This indicates that the agents could ease the administrative burden by removing the need to learn software development skills.
Download
Paper Nr: 257
Title:

Optical Character Recognition Based-On System for Automated Software Testing
Authors:

D. Abbas and J. I. Olszewska
Abstract: The paper presents the development and deployment of an artificial intelligence (AI) test automation framework that allows testers to more fluidly develop scripts and carry out their day-to-day tasks. In particular, the framework aims to speed up the test automation process by enabling its users to locate elements on a webpage through the use of template-matching-based image recognition as well as optical character recognition (OCR). Indeed, test automation specialists spend much of their time creating page-object models (POMs), where they capture elements on the screen via complex locators such as cascading style sheet (CSS) or XPath. However, when webpages are updated or elements are moved around, locators become void, eventually pointing to nothing unless written in such a dynamic way as to prevent this. This heavily relies on developers providing meaningful tags to elements that they can then be located by, whereas with the introduction of an image recognition engine in our AI framework, this tedious and long-winded approach has been be shortened.
Download

Area 3 - Information Systems Analysis and Specification

Full Papers
Paper Nr: 86
Title:

Balancing Autonomy and Control: An Adaptive Approach for Security Governance in Large-Scale Agile Development
Authors:

Sascha Nägele, Nathalie Schenk, Nico Fechtner and Florian Matthes
Abstract: Companies are increasingly adopting agile methods at scale, revealing a challenge in balancing team autonomy and organizational control. To address this challenge, we propose an adaptive approach for security governance in large-scale agile software development, based on design science research and expert interviews. In total, we carried out 28 interviews with 18 experts from 15 companies. Our resulting approach includes a generic organizational setup of security-related roles, a team autonomy assessment model, and an adaptive collaboration model. The model assigns activities to roles and determines their frequency based on team autonomy, balancing the autonomy-control tension while ensuring compliance. Although framework-agnostic, we applied our approach to existing scaling agile frameworks to demonstrate its applicability. Our evaluation indicates that the approach addresses a significant problem area and provides valuable guidance for incorporating security into scaled agile environments. While the primary focus is on security governance, our insights may be transferable to other cross-cutting concerns.
Download
Paper Nr: 130
Title:

Project Management in Large-Scale with International Settings: Challenges Faced with Multiculturalism
Authors:

Luana Mendes, Vinicius Faria and Cristiano Maciel
Abstract: Managing large-scale international research projects involving multicultural teams poses multifaceted challenges requiring a nuanced and adaptive project management approach. This article explores the strategies employed to manage a multifaceted research and innovation initiative within the "ELLAS - Equality in Leadership for Latin American STEM" Network, focusing on gender gap reduction in STEM fields across Latin America. Combining elements from PMBOK® and agile methodologies. The project integrates a diverse array of institutions and stakeholders from multiple countries. Our study delves into the project’s objectives, team structures, deliverables, and the unique challenges encountered, including multilingual communication, bureaucratic complexities, managing multiple currencies, and ensuring data quality. We detail solutions applied to navigate these challenges, such as tailored communication strategies, hybrid project management methodologies, and proactive stakeholder engagement. By delineating these challenges and solutions, this article brings insights into managing complex international research projects while fostering cultural diversity and achieving project objectives.
Download
Paper Nr: 133
Title:

On the Integration of Privacy-Enhancing Technologies in the Process of Software Engineering
Authors:

Alexandra Klymenko, Stephen Meisenbacher, Luca Favaro and Florian Matthes
Abstract: The class of technologies known as Privacy-Enhancing Technologies (PETs) has been receiving rising attention in the academic sphere. In practice, however, the adoption of such technologies remains low. Beyond the actual implementation of a PET, it is not clear where along the process of software engineering PETs should be considered, and which activities must take place to facilitate their implementation. In this light, we aim to investigate the placement of PETs in the software engineering process, specifically from the perspective of privacy requirements engineering. To do this, we conduct a systematic literature review and interview 10 privacy experts, exploring the integration of PETs into the software engineering process, as well as identifying associated challenges along with their potential solutions. We systematize our findings in a unified process diagram that illustrates the roles and activities involved in the implementation of PETs in software systems. In addition, we map the identified solution concepts to the diagram, highlighting which stages of the software engineering process are vital in tackling the corresponding challenges and supporting the adoption of PETs.
Download
Paper Nr: 153
Title:

Ontology to Define Sizing Screw Joints for Mechanical Engineering Applications
Authors:

Henrique Priebe, Gustavo R. Ramos and Vinícius Maran
Abstract: The sizing of bolted joints is crucial for mechanical projects that will experience working loads. This requires careful analysis during the design specification phase, where materials, manufacturing processes, components, and layout are determined. This analysis is time-consuming since most of it is done manually using analytical calculations, as there are limited computational tools available. Developing such tools requires a formal representation of knowledge that algorithms can process. The purpose of this work is to create a conceptual model, using ontologies, to represent the information found in specialized literature about bolts and bolted joints. By doing so, it becomes possible to automate calculations for bolt stiffness and stiffness of joint elements. The ontology was built following the UPON methodology, which involves extracting domain terms to guide knowledge modeling. The resulting ontology provides a formal and explicit representation of statically loaded bolted joints in an axial direction. It also has the capability to answer predetermined competence questions, indicating that it can be used by software applications to process information efficiently.
Download
Paper Nr: 155
Title:

Software Engineers Engagement and Job Satisfaction: A Survey with Practitioners Working Remotely in a Public Organization
Authors:

Lidiany Cerqueira, Lourene L. Nunes, Viviane Malheiros, Renan Guerra, Beatriz Santana, Rodrigo Spínola, Manoel Mendonça and José M. Santos
Abstract: Context: Work engagement is related to a positive fulfilling work-related mental state. Job satisfaction refers to how professionals are satisfied with their work. Measuring work engagement and job satisfaction can help organizations to foster employee productivity, as they are related. Objective: This study aims to analyze the work engagement and job satisfaction of software practitioners working in remote environment in a public organization. Method: We assess the engagement and job satisfaction of software professionals at a large governmental software organization. We surveyed a group of 148 employees and performed a quantitative and qualitative analysis of the responses. Results: The respondents reported good level of engagement and job satisfaction, 63% of them would recommend their team to a friend. The survey also reveals that career development, psychological safety, team, management and rewards, benefits, meeting planning, and social interactions are the factors that most affect the satisfaction of software professionals. Conclusion: The results of this study can help software organizations in fostering workplace improvement and satisfaction of software development teams. For researchers, results provide a grounded view of work engagement and job satisfaction, guiding new research efforts aligned with the demands and current context as experienced by practitioners. For practitioners, the identified factors provide empirical reference for improving work environments. We summarized them in a cheat sheet frame.
Download
Paper Nr: 201
Title:

A Rule-Based Log Analysis Approach for State-Machine Governed Systems
Authors:

Jeroen Zwysen, Felicien Ihirwe, Ken Vanherpen, Maarten Vergouwe, Umut Caliskan and Davy Maes
Abstract: Logs are used in programming for various purposes, ranging from failure analysis to software comprehension. However, the processing of logs is hindered by the lack of structure in the logs, the required domain knowledge for interpretation, and a lack of tooling. In this paper, a novel approach that includes structured log generation and rule-based log analysis is presented. Targeting state machine-governed systems, the approach relies on developers’ knowledge during design time to allow hierarchical grouping of logs and standard visualization of the logs during the analysis. This allows automated failure diagnosis and localization without full system-wide domain knowledge as well as providing a historical context of the system during a failure event. To better evaluate the effectiveness of the approach, two use cases, namely a Virtual Coffee Machine (VCM) and an Automated Mobile Robot (AMR) are showcased and analyzed.
Download
Paper Nr: 247
Title:

CarbonSECO for Livestock: A Service Suite to Help in Carbon Emission Decisions
Authors:

Pedro A. Silva, Regina Braga, José M. David, Valdemar G. Neto, Wagner Arbex and Victor Stroele
Abstract: Global concerns about agriculture’s impact, mainly related to the livestock enteric fermentation producing methane Greenhouse Gas (GHG) emissions, demand solutions to mitigate these impacts. The CarbonSECO platform, tailored for carbon credit generation in Brazilian rural areas, responds to the popularity of carbon credits to offset GHG emissions. However, additional solutions related to GHG need to be conceived. This article extends the CarbonSECO platform, focusing on quantifying, monitoring, and controlling carbon emissions from livestock enteric fermentation. Intelligent techniques, including ontologies and machine learning, provide emissions management solutions for assessing and managing the environmental impact of livestock farming. These techniques address the research question of mitigating carbon emissions from Brazilian dairy farming. The article explores strategies, reviews related works, and proposes platform extensions. A feasibility study using data from an intelligent farm showcases the platform’s ability to predict and assess changes for carbon emission reduction. As a result, this work enhances the CarbonSECO platform, offering emissions management for dairy farming. Integrating ontologies and machine learning can promote standardization, aiding property owners in better planning.
Download
Paper Nr: 253
Title:

Greener Information Systems for Product Configuration Management: Towards Adaptation to Sustainability Requirements
Authors:

Anders Jakobsen, Torben Tambo and Maja Kadenic
Abstract: The purpose of this paper is to shed light on the need to reconceptualize the dimension of product life-cycle management systems related to product configuration to embrace data of the specific sustainability impact of the configuration choices. While this is very much related to physical products, the information systems dimension is fundamental to include to model, decide, document, trace and review sustainability of products. This paper is based on a longitudinal case study along with a comprehensive literature review. Key findings related to the isolation of product configuration systems as key determinants for specific sustainability in a governed and traceable form. These systems do largely not cover sustainability as of today: A redesign is needed. A research agenda is outlined combining sustainability-thinking with socio-technical design. A proposal for the design is presented using a multi-level, multi-tier approach to Product Configuration Systems. The process has major implications around in the industry as legislators are mandating extensive documentation for specific choices and documentation of the sustainability impact of physical products.
Download
Short Papers
Paper Nr: 21
Title:

Requirements Engineering for Continuous Queries on IoRT Data: A Case Study in Agricultural Autonomous Robots Monitoring
Authors:

Leandro Antonelli, Hassan Badir, Houssam Bazza, Sandro Bimonte and Stefano Rizzi
Abstract: The Internet of Robotic Things (IoRT) is an extension of the Internet of Things, where intelligent mobile devices acquire sensor data and physically act in the environment. IoRT devices produce huge data streams, typically analyzed using continuous queries. We propose an approach to engineer requirements about continuous queries over IoRT data. Our proposal is specifically devised for end-users not skilled in IT and relies, for requirements elicitation, on spreadsheet-like templates called stream tables. Requirements analysis uses a novel UML profile, while requirements specification and validation rely on a fast prototyping tool so as to allow end-users to define continuous queries by themselves and validate them via web-based prototyping. Non-functional requirements are taken into account as well, in the form of available technological resources and data sources, and used for requirements validation. The results of some preliminary tests made with some real users suggest that stream tables are a valuable instrument for the engineering of continuous queries, and that fast prototyping is an effective support to the specification and validation steps.
Download
Paper Nr: 48
Title:

Model-Based Auto-Commissioning of Building Control Systems
Authors:

Philipp Zech, Emanuele Goldin, Sascha Hammes, David-Geisler Moroder, Rainer Pfluger and Ruth Breu
Abstract: Digital twins are valuable instruments for model-based design, commissioning, and operation, with significant applicability potential in the construction industry. Whereas with Building Information Modeling (BIM) a standard for the representation of building models has been established, these models lack (i) modeling support for building control systems, and (ii) tool-based automation support for model-based auto-commissioning of building automation systems, an instrumental factor in putting a digital twin in operation. In this paper, we present a domain-specific language (DSL), its modeling methodology, and tool support to augment and condition BIM models for auto-commissioning. Preliminary results from an early prototype evaluation using the Technology Acceptance Model demonstrate the feasibility of our proposal in contributing to the improvement of building operations by facilitating auto-commissioning of building control systems and subsequent commissioning of digital twins.
Download
Paper Nr: 54
Title:

Testing on Dynamically Adaptive Systems: Challenges and Trends
Authors:

Isabely N. Costa, Ismayle S. Santos and Rossana C. Andrade
Abstract: Dynamically Adaptive Systems (DAS) are systems capable of modifying themselves automatically according to the surrounding environment. Traditional testing approaches are ineffective for these systems due to their dynamic aspects, making fault detection complex. Although various testing approaches have been proposed for DASs, there is no up-to-date overview of the approaches, challenges, and trends. This research therefore presents the results of a systematic literature review to identify the challenges, approaches and trends in testing dynamically adaptable systems. For this objective, 25 articles between 2020 and 2023 were analyzed to answer our research questions. As a result, approaches and their characteristics were identified, such as what type of system they can be applied to, what activity is included in the testing process, and at what level of testing. We also highlighted challenges that are still being faced and trends in testing dynamically adaptive systems. For a more in-depth analysis of the results related to the challenges, grounded theory procedures were applied to organize them and encourage future research that seeks to overcome and mitigate them.
Download
Paper Nr: 58
Title:

Cross-Domain Classification of Domain Entities into Top-Level Ontology Concepts Using BERT: A Study Case on the BFO Domain Ontologies
Authors:

Alcides Lopes, Joel Carbonera, Nicolau Santos, Fabricio Rodrigues, Luan Garcia and Mara Abel
Abstract: Classifying domain entities into top-level ontology concepts using informal definitions remains an active research area with several open questions. One of these questions pertains to the quality of proposed pipelines employing language models for classifying informal definitions when training and testing samples are from different knowledge domains. This can introduce challenges due to varying vocabularies across domains or the potential for an entity to belong to different top-level concepts based on its domain. In this study, we present a study case where terms and informal definitions are extracted from 81 domain ontologies organized into 12 knowledge domains. We investigate the performance of a pipeline that utilizes the BERT language model for classifying domain entities into top-level concepts within a cross-domain classification scenario. Additionally, we explore various pipeline setups for input, preprocessing, and training steps. Our optimal classifier setup employs an unbalanced training methodology, no text preprocessing, and the concatenation of terms and informal definitions as input. Furthermore, we demonstrate that BERT yields promising results in classifying domain entities into top-level concepts within a cross-domain classification scenario.
Download
Paper Nr: 65
Title:

Squeezing the Lemon: Using Accident Analysis for Recommendations to Improve the Resilience of Telecommunications Organizations
Authors:

Hans A. Wienen, Faiza A. Bukhsh, Eelco Vriezekolk, Luís F. Pires and Roel J. Wieringa
Abstract: Telecommunications networks form critical infrastructure, since accidents in these networks can severely impact the functioning of society. Structured accident analysis methods can help draw lessons from accidents, giving valuable information to improve the resilience of telecommunications networks. In this paper, we introduce a method (TRAM) for accident analysis in the Telecommunication domain by improving AcciMap, which is a popular method for analyzing accidents. The improvements have made AcciMap more efficient and instructive by explicitly identifying ICT aspects of the accidents, extending the method to support the evaluation of crisis organizations and introducing additional notation for feedback loops. This resulted in TRAM, a method with a 25% improved efficiency over AcciMap, while also addressing ICT aspects, leading to concrete actionable results that can help telecommunication organizations grow more resilient.
Download
Paper Nr: 70
Title:

Can a Chatbot Support Exploratory Software Testing? Preliminary Results
Authors:

Rubens Copche, Yohan Duarte, Vinicius Durelli, Marcelo M. Eler and Andre T. Endo
Abstract: Tests executed by human testers are still widely used in practice and fill the gap left by limitations of automated approaches. Among the human-centered approaches, exploratory testing is the de facto approach in agile teams. Although it is focused on the expertise and creativity of the tester, the activity of exploratory testing may benefit from support provided by an automated agent that interacts with human testers. We set out to develop a chatbot named BotExpTest, specifically designed to assist testers in conducting exploratory tests of software applications. We implemented BotExpTest on top of the instant messaging social platform Discord; this version includes functionalities to report bugs and issues, time management of test sessions, guidelines for app testing, and presentation of exploratory testing strategies. To assess BotExpTest, we conducted a user study with six software engineering professionals. They carried out two sessions performing exploratory tests along with BotExpTest. Participants revealed bugs and found the experience to interact with the chatbot positive. Our analyses indicate that chatbot-enabled exploratory testing may be as effective as similar approaches and help testers to uncover different bugs. Bots are shown to be valuable resources for Software Engineering, and initiatives like BotExpTest may help to improve the effectiveness of testing activities like exploratory testing.
Download
Paper Nr: 87
Title:

Deep Convolutional Neural Network and Character Level Embedding for DGA Detection
Authors:

João R. Gregório, Adriano M. Cansian, Leandro A. Neves and Denis P. Salvadeo
Abstract: Domain generation algorithms (DGA) are algorithms that generate domain names commonly used by botnets and malware to maintain and obfuscate communication between a botclient and command and control (C2) servers. In this work, a method is proposed to detect DGAs based on the classification of short texts, highlighting the use of character-level embedding in the neural network input to obtain meta-features related to the morphology of domain names. A convolutional neural network structure has been used to extract new meta-features from the vectors provided by the embedding layer. Furthermore, relu layers have been used to zero out all non-positive values, and maxpooling layers to analyze specific parts of the obtained meta-features. The tests have been carried out using the Majestic Million dataset for examples of legitimate domains and the NetLab360 dataset for examples of DGA domains, composed of around 56 DGA families. The results obtained have an average accuracy of 99.12% and a precision rate of 99.33%. This work contributes with a natural language processing (NLP) approach to DGA detection, presents the impact of using character-level embedding, relu and maxpooling on the results obtained, and a DGA detection model based on deep neural networks, without feature engineering, with competitive metrics.
Download
Paper Nr: 123
Title:

Micro Frontend-Based Development: Concepts, Motivations, Implementation Principles, and an Experience Report
Authors:

Fernando D. Moraes, Gabriel N. Campos, Nathalia D. Almeida and Frank J. Affonso
Abstract: Micro frontend is an architectural style that enables us to build large software systems by combining independent micro applications, which can boost different aspects related to the development (e.g., innovation, continuous software delivery), besides increasing the flexibility and scalability of the final application itself. Although there are numerous benefits related to this architectural style, some companies are still hesitant to adopt development based on micro frontends because of a lack of knowledge about concepts, development approaches, architectural models, and organizational aspects of the company. This paper presents the results of a Systematic Mapping Study (SMS) on micro frontends based on 16 studies. The results were synthesized in an important overview that addressed concepts, aspects related to development (i.e., development approaches, architectural models, and company organization), and micro frontend trade-offs based on three scenarios. Next, we present a case study on an inventory control application based on the knowledge of this SMS, analyzing the development under three approaches (i.e., Build-time, Frameworkless, and Framework-based). As result, we observed our paper has a good perspective to contribute efficiently to the micro frontend domain by providing an overview of this research area and an experience report for researchers and practitioners.
Download
Paper Nr: 146
Title:

Can Personality Types Be Blamed for Code Smells?
Authors:

Manoel Valerio D. Silveira Neto, Andreia Malucelli and Sheila Reinehr
Abstract: The term code smell refers to sections of code that are not technically incorrect, do not prevent the software from functioning, but affect its quality. Code smells are considered a form of technical debt (TD). This study investigated the relationship between the personality types of software developers and the presence of code smells, which indicate potential problems in source code. Using the Myers-Briggs Type Indicator (MBTI) to classify personalities, the study examines whether specific profiles are more associated with creating or removing code smells. The goal is to assist software project managers in allocating tasks for refactoring and development. The research does not find a statistically significant correlation between the developer’s personality and the creation of code smells. Still, it suggests that the Consul personality type (ESFJ) shows a greater tendency to resolve code smells. The study also highlights the importance of considering human factors such as personality types in software development to improve product quality.
Download
Paper Nr: 154
Title:

Effective People Management Practices for Software Project Success
Authors:

Marcelo F. Burkard and Lisandra M. Fontoura
Abstract: Research on people management practices is crucial because they significantly influence the results of software projects, help improve decision compliance, and maintain a qualified workforce. However, there is a tendency for managers to rely on personal experience rather than evidence-based knowledge when implementing people management practices. Compiling good practices can assist managers in implementing people management practices, reducing resistance, and, at the same time, collecting indicators that can be used to evaluate the effectiveness of practices. This work documents good practices that can support people management in software projects based on the compilation of practices. To this end, we carried out a systematic literature review (SLR) to identify common problems in people management in software projects and effective practices to resolve them. Initially, SLR returned 2495 unduplicated primary studies. After a detailed analysis, 63 studies were selected and organized into nine problem categories and sixteen practices. Through a survey, these practices were validated by 31 software professionals, allowing them to be classified according to the general relevance of the practice and to resolve each associated problem. The findings reveal the predominance of interpersonal skills (soft skills) over technical skills (hard skills) and emphasize the importance of practices such as continuous feedback, open communication, and transparent management.
Download
Paper Nr: 158
Title:

Investigating Entry-Level Software Project Managers’ Skills and Responsibilities: An Empirical Analysis of LinkedIn Job Ads
Authors:

Clara Berenguer, Sávio Freire, Manoel Mendonça and Rodrigo Spínola
Abstract: Context. Project managers play a central role in software development projects. Knowing the skills required and responsibilities expected from entry-level software project managers can help those starting a job search and organizations that want to articulate their staffing needs clearly. Aims. To investigate the required skills and expected responsibilities prevalent in the job market for entry-level software project managers. Method. This work collects and analyzes, qualitatively and quantitatively, 50 online job advertisements from the LinkedIn Jobs platform. Results. Overall, organizations look for professionals with a vast list of skills and able to address several job responsibilities. The most expected responsibilities are planning, organizing, and coordinating team activities, establish a good communication with the client, and analyze and communicate project risks. The most required skills are communication, technical knowledge, and planning/managing. Both hard and soft skills are expected. However, soft skills are slightly prevalent. Conclusion. Job ads are a valuable source to gain insights into current job market trends and project management role expectations for professionals, organizations, and researchers.
Download
Paper Nr: 196
Title:

Towards Legal Interoperability in International Data Spaces
Authors:

Victor B. Jales de Oliveira, Patrício A. Silva and João R. Moreira
Abstract: The value of data exchange is indubitably a thriving approach, however, it must be conducted in a safe and sovereign space, avoiding the loss of control, and data misusage. The International Data Spaces (IDS) is supposed to be a trusted environment, in which companies could share sensitive data upholding data sovereignty. Thus, mitigating the risk of losing industrial secrets and further threats to competition. Along with the mentioned two foundations of IDS, its architecture allows a free contract endorsement, on which, companies may negotiate their policies and governing laws. A service contract should be able to unambiguously represent all involved policies, leaving no breach for subjectivity. Another important aspect of IDS is to follow the Findable, Accessible, Interoperable, and Reusable (FAIR) principles. In particular, we focus on the Legal Interoperability. As one of the proposed interoperability layers (intended by the European Interoperability Framework), Legal Interoperability is proposed as the capability of companies from different countries (under different governing laws) to cooperate. This paper provides a research agenda and presents prior results of the proposed methodologies, addressing how to resolve legal interoperability issues before establishing IDS legal agreements. It takes a Design Science perspective for problem decomposition into specific issues, triangulation of research methods, and projection of a solution space.
Download
Paper Nr: 205
Title:

Exploring the Pros and Cons of Monolithic Applications versus Microservices
Authors:

Daniel S. Krug, Rafael Chanin and Afonso Sales
Abstract: Microservices architecture emerged as an alternative to monolith architecture. With monoliths, applications are developed in entire blocks that communicate internally, manage their data usually in a single database, and each new feature demands the deployment of the application as a whole. On the other hand, microser-vices splits the application into smaller blocks with unique responsibilities, using lightweight communication mechanisms and managing their own data. This new architecture has several advantages, but it also has some disadvantages. From the understanding of these advantages and disadvantages, the main goal of this research is to identify how the two architectures have been used in professional practices. As well as how the academy can help in the understanding that if the microservices architecture entails a longer development time for the applications, in order to understand when the decision for an architecture may be more appropriate with another in software development. To fulfill the explained objective, it is intended to carry out a systematic study with the snowballing technique, research in the grey literature and a survey with specialists.
Download
Paper Nr: 222
Title:

Challenges in Reverse Engineering of C++ to UML
Authors:

Ansgar Radermacher, Marcos D. Fabro, Shebli Anvar and Frédéric Chateau
Abstract: Model-driven engineering provides several advantages compared to a direct manual implementation of a system. In reverse-engineering applications, an existing code basis needs to be imported into the modeling language. However, there is an abstraction gap between the programming language (C++) and the modeling language, in our case UML. This gap implies that the model obtained via reverse engineering is a model that directly mirrors the object-oriented implementation structures and does not use higher-level modeling mechanisms such as component-based concepts or state-machines. In addition, some concepts of the implementation languages can not be expressed in UML, such as advanced templates. Therefore, new systems are often either developed from scratch or model-driven approaches are not applied. The latter has become more attractive recently, as IDEs offer powerful refactoring mechanisms and AI based code completion - model-driven approaches need to catch up with respect to AI support to remain competitive. We present a set of challenges, based on examples, that need to be handled when reverse engineering C++ code. We describe how we handle them by improving reverse engineering capabilities of an existing tool.
Download
Paper Nr: 233
Title:

Forecasting of Key Performance Indicators Based on Transformer Model
Authors:

Claudia Diamantini, Tarique Khan, Alex Mircoli and Domenico Potena
Abstract: Key performance indicators (KPIs) express the company’s strategy and vision in terms of goals and enable alignment with stakeholder expectations. In business intelligence, forecasting KPIs is pivotal for strategic decision-making. For this reason, in this work we focus on forecasting KPIs. We built a transformer model architecture that outperforms conventional models like Multi-Layer Perceptrons (MLP), Long Short-Term Memory (LSTM) networks, Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN) in KPI forecasting over the Rossmann Store, supermarket 1, and 2 datasets. Our results highlight the revolutionary potential of using cutting-edge deep learning models such as the Transformer.
Download
Paper Nr: 245
Title:

Towards an Eco-Gamification Platform to Engage Consumers in the Textile & Clothing Circular Economy
Authors:

Estrela F. Cruz, David Braga, Diogo Assunção, Pedro M. Faria and A. M. Rosado da Cruz
Abstract: Nowadays people are buying too many clothes. This results in the overexploitation of the Earth’s natural resources and an increase in waste that is not sustainable. It is necessary to consume fewer new resources, and alternatively reuse more, repair if necessary, share and donate more, and at the end of the garment life it is necessary to transform it to be used as raw material in the creation of new products and in this way contribute to the circular economy. This article proposes the creation of a gamified platform that engages consumers, brands and other participants in more sustainable practices at the same time that helps consumers to manage their clothing closet. It also allows consumers to manage other closets, such as their children’s closet or their home textiles. The main goal is to encourage consumers to adopt more sustainable practices with regard to clothing and home textiles. The proposed platform allows to share and donate clothes through the creation of sharing/donation groups. The main objective is to contribute to the transition of the textile and clothing sector towards the circular economy, making it a more sustainable sector.
Download
Paper Nr: 143
Title:

A Method for Marketing and Sales Optimization of Enterprise Software Applications
Authors:

Dana van Uitert, Marijn Smit, Wim van den Brandt and Sjaak Brinkkemper
Abstract: Marketing and Sales departments are closely related but there is also underperformance within this collaboration. Multiple strategies exist to improve collaboration or integration between the two. This paper focuses on the Discover Your Potentials method which uses a combination of different models to improve business performance in enterprise software companies. A qualitative case study approach was used to find out how successful the method is in designing marketing and sales potentials. The paper discusses how the method was executed at a company developing and selling accounting software extensions. Based on this study some preliminary findings and conclusions can be formulated.
Download
Paper Nr: 163
Title:

Validation and Clarification of Critical Success Factors of DevOps Processes
Authors:

Michiel van Belzen, Jos Trienekens and Rob Kusters
Abstract: Context: Critical Success Factors (CSFs) may contribute to solve challenges regarding Continuous Integration, Continuous Delivery and Continuous Deployment processes (CI/CD-processes). Prior research found some CSFs related to CI/CD and aspects of DevOps, but they are limited regarding validation, clarification and comprehensiveness. Objective: This study aims to contribute to the success of CI/CD-processes by showing and clarifying which CSFs determine the success of CI/CD-processes. Method: A three-phase process was followed. In the first phase, we conducted a systematic literature review in which we identified 144 potential CSFs. In the second phase, we classified the CSFs found into nineteen potential CSFs. Finally, we conducted a multiple case study with the following objectives: (1) to find examples of application to show that the potential CSFs were recognized by experts in the field, (2) to use the examples to validate the potential CSFs and show how the CSFs could be operationalized, and (3) to clarify why the validated CSFs are important to the success of CI/CD-processes. Results: Our main contribution to theory is a validated and structured model of nineteen clarified CSFs of CI/CD-processes, which were understood, recognized and grounded in practice by examples and clarifications on the importance of CSFs. Conclusions: Presenting a comprehensive model of CSFs, it appears that we achieved consensus regarding CSFs of CI/CD-processes in literature. In addition, IT-organizations could apply this model of CSFs to take steps towards successful results of CI/CD-processes.
Download
Paper Nr: 203
Title:

Challenges of Trustworthy of Digital Evidence and Its Chain of Custody on Cloud Computing Environment: A Systematic Review
Authors:

Lucien R. Lucien
Abstract: For reliable digital evidence to be admitted in a court of law, it is important to apply scientifically proven digital forensic investigation techniques to corroborate a suspected security incident. Mainly, traditional digital forensics techniques focus on computer desktops and servers. However, recent advances in usage of cloud computing environments increased the need for the application of digital forensic investigation techniques to their infrastructure, that has some particularities, such as multi-jurisdictions storage, improper handling by third parties, high level of volatility, etc. In this paper, we perform a systematic review about the challenges of thustworthy of digital evidence and its chain of custody (CoC) on cloud computing environment. The literature search yielded 32 articles that met the study criteria. It resulted in mapping the main challenges found in the literature when applying existing approaches to increase the admissibility in courts of digital evidence collected on cloud computing environment. Furthermore, this work aims to update the systematic research regarding this subject covering the period of 2020 to 2023.
Download
Paper Nr: 204
Title:

Development of a Ship Mooring Inspection Winch Tool with Extended Reality
Authors:

Wagner Aparecido de Oliveira, Saul D. Silva and Adrielle C. Santana
Abstract: Over the years, the interactivity between the real world and the virtual world has been increasing, making it attractive for engineers to increasingly seek new tools for application in industrial areas Practices that reconcile the safety of people and assets, adds strength to investments. The focus here is a solution for working with winches for ship mooring systems. The mooring system used to moor ships at a pier is made up of methods characterized using combinations of hooks, bollards, and winches. Through the instrumentation on board this equipment, information is obtained to analyze the stabilization of the mooring. The main parts of the winch system consist of the sensing and measuring system, cable tension measurement system, electrical and the mechanical system. This work proposes the development of an extended reality application for training maintenance teams. The application will enable the trained team to interact with the equipment’s functionalities and information in a virtual and safe way, thus ensuring they have access to the operation, fault diagnosis with simulations of problem solutions in a virtual reality environment. Is still expected to be added, in future developments, a of virtual interfaces for remote equipment operations increasing the speed of fault identification.
Download
Paper Nr: 213
Title:

Data Discovery and Indexing for Semi-Structured Scientific Data
Authors:

Kaushik Jagini, Yifan Zhang, Yichen Guo, Julian Goddy, Dale Stansberry, Joshua Agar and Jeff Heflin
Abstract: There is a need for powerful, user-friendly tools for scientific data management and discovery. We present an architecture based on DataFed and Elasticsearch that allows scientists to easily share data they produce and a novel interface that allows other scientists to easily discover data of interest. This interface supports summary-level information about a collection of datasets that can be easily refined using schema-free search. We extend the recent idea of cell-centric search to semi-structured data, describe the architecture of the system, present a use case from the context of materials science, and evaluate the efficacy of the system.
Download
Paper Nr: 243
Title:

Data Quality Assessment for the Textile and Clothing Value-Chain Digital Product Passport
Authors:

A. M. Rosado da Cruz, Pedro Silva, Sérgio Serra, Rodrigo Rodrigues, Pedro Pinto and Estrela F. Cruz
Abstract: The Textile and Clothing (T&C) industrial sector is transforming to become more sustainable and in line with the directives of the European Union. Therefore, to become more transparent and gain consumer trust, some projects present proposals to implement the traceability of T&C products. However, this sector has a very large and diverse value chain that involves many types of industries that are typically spread throughout the world. Furthermore, a previously developed project to implement traceability on the value chain reveals that the involved companies have different levels of digital maturity and, among those with the same level of maturity, different digital platforms are used. Consequently, some values submitted for a T&C traceability platform may be collected automatically, while others have to be manually inserted. This makes it necessary to create a tool for validating the data values submitted to the traceability platform, which can be integrated into the different organizational tools so that the data can be validated homogeneously. After summarizing the relevant and contextualizing facts about the T&C value chain, and reviewing the data quality assurance mechanisms, this paper proposes a software service for validating data values of metrics being traced across the T&C value chain, that integrates the Digital Product Passport of T&C items. Associated with the validation service, an admin platform for configuring the service for each metric is also proposed.
Download
Paper Nr: 260
Title:

Professional Insights into Benefits and Limitations of Implementing MLOps Principles
Authors:

Gabriel Araujo, Marcos Kalinowski, Markus Endler and Fabio Calefato
Abstract: Context: Machine Learning Operations (MLOps) has emerged as a set of practices that combines development, testing, and operations to deploy and maintain machine learning applications. Objective: In this paper, we assess the benefits and limitations of using the MLOps principles in online supervised learning. Method: We conducted two focus group sessions on the benefits and limitations of applying MLOps principles for online machine learning applications with six experienced machine learning developers. Results: The focus group revealed that machine learning developers see many benefits of using MLOps principles but also that these do not apply to all the projects they worked on. According to experts, this investment tends to pay off for larger applications with continuous deployment that require well-prepared automated processes. However, for initial versions of machine learning applications, the effort taken to implement the principles could enlarge the project’s scope and increase the time needed to deploy a first version to production. The discussion brought up that most of the benefits are related to avoiding error-prone manual steps, enabling to restore the application to a previous state, and having a robust continuous automated deployment pipeline. Conclusions: It is important to balance the trade-offs of investing time and effort in implementing the MLOps principles considering the scope and needs of the project, favoring such investments for larger applications with continuous model deployment requirements.
Download
Paper Nr: 263
Title:

A Distributed Processing Architecture for Disease Spread Analysis in the PDSA-RS Platform
Authors:

Denilson S. Ebling, Felipe Machado, Glenio Descovi, Nicolas Cardenas, Gustavo Machado, Vinicius Maran and Alencar Machado
Abstract: In today’s world, machine learning systems have permeated various domains, from object detection to disease spread prediction, playing pivotal roles in decision-making processes. Amid the COVID-19 pandemic, the utilization of machine learning methods like artificial neural networks and LSTM networks has significantly enhanced forecasting accuracy for disease outbreaks. This paper delves into the development of an intelligent system proposed by Cardenas et al. (2022a), focusing on simulating disease spread in animals and facilitating control measures through a stochastic model. Leveraging Docker containers for deployment, this system offers valuable insights for public health interventions, enabling swift responses to disease outbreaks. The primary objective of this work is to provide veterinarians with a user-friendly tool that integrates a stochastic model through an intuitive interface, aiding in critical decision-making processes in a scalable manner. The paper outlines the background of the stochastic model, introduces the proposed system for integrating and addressing the identified problem, presents an evaluation scenario to validate the system’s efficacy, and concludes with insights drawn from this research endeavor.
Download

Area 4 - Software Agents and Internet Computing

Full Papers
Paper Nr: 93
Title:

EQNet: A Post-Processing Approach to Manage Popularity Bias in Collaborative Filter Recommender Systems
Authors:

Gabriel V. Machado, Wladmir C. Brandão and Humberto T. Marques-Neto
Abstract: Recommendation systems play a pivotal role in digital platforms, facilitating novel user experiences by effectively sorting and presenting items that align with their preferences. However, these systems often suffer from popularity bias, a phenomenon characterized by the algorithm’s inclination to favor a few popular items, resulting in the under-representation of the vast majority of items. Addressing this bias and enhancing the recommendation of long-tail items is of utmost importance. In this paper, we propose the EQNet, a re-ranking approach designed to mitigate popularity bias and improve the recommendation quality of an SVD-based recommendation system. EQNet leverages PageRank or Popularity Count outputs to re-rank items, and its effectiveness is evaluated using four metrics: average popularity, percentage of long-tailed items, coverage of long-tailed items, and recommendation quality. We incorporate the widely recognized bias mitigation algorithm FA*IR into our experimentation to establish a robust baseline. By comparing the performance of EQNet against this state-of-the-art approach, we show the efficiency of EQNet and highlight its potential to enhance existing methods for mitigating popularity bias.
Download
Paper Nr: 142
Title:

A Systematic Mapping on Software Aging and Rejuvenation Prediction Models in Edge, Fog and Cloud Architectures
Authors:

Paulo A. Costa, Edward M. Ordonez and Jean C. Teixeira de Araujo
Abstract: This article presents a Systematic Literature Mapping (SLM), related to software aging and rejuvenation prediction models. The study highlights the importance of these models, due to the high cost of software or service downtime in IT datacenter environments. To mitigate this impact and seek greater reliability and availability of applications and services, software aging prediction and proactive rejuvenation are significant research topics in the area of Software Aging and Rejuvenation (SAR). Costs are potentially higher when rejuvenation actions are not scheduled. Various prediction models have been proposed for over twenty-five years, with the aim of helping to find the ideal moment for rejuvenation, in order to optimize the availability of services, reduce downtime and, consequently, the cost. However, the scope of this study was limited to a survey of the last fifteen years of models with a measurement-based prediction strategy. These models involve monitoring and collecting data on resource consumption over time, from a running computer system. The collected data is used to adjust and validate the model, allowing the prediction of the precise moment of the aging phenomenon and the consequent rejuvenation action of the software. In addition to providing a baseline from the compiled prediction models, identifying gaps that could encourage future research, particularly in the areas of machine learning or deep learning, the research also contributed to clarifying that hybrid algorithms based on Long Short-Term Memory (LSTM) are currently situated at the highest level of prediction models for software aging, with recent highlights for two variants: the Gated Recurrent Unit (GRU) and the Bidirectional Long Short Term Memory (BiLSTM). Objectively, in response to the research questions, the article also contributes by presenting, through tables and graphs, trends and consensus among researchers regarding the evolution of prediction models.
Download
Paper Nr: 252
Title:

An Extension of Orbslam for Mobile Robot Using Lidar and Monocular Camera Data for SLAM Without Odometry
Authors:

Rodrigo L. Santos, Mateus C. Silva and Ricardo R. Oliveira
Abstract: Mobile autonomous robots require accurate maps to navigate and make informed decisions in real-time. The SLAM (Simultaneous Localization and Mapping) technique allows robots to build maps while they move. However, SLAM can be challenging in complex or dynamic environments. This study presents a mobile autonomous robot named Scramble, which uses SLAM based on the fusion of data from two sensors: a RPLIDAR A1m8 LiDAR and an RGB camera. How to improve the accuracy of mapping, trajectory planning, and obstacle detection of mobile autonomous robots using data fusion? In this paper, we show that the fusion of visual and depth data significantly improves the accuracy of mapping, trajectory planning, and obstacle detection of mobile autonomous robots. This study contributes to the advancement of autonomous robot navigation by introducing a data-fusion-based approach to SLAM. Mobile autonomous robots are used in a variety of applications, including package delivery, cleaning, and inspection. The development of more robust and accurate SLAM algorithms is essential for the use of these robots in challenging environments.
Download
Short Papers
Paper Nr: 43
Title:

Exploring Strategies to Mitigate Cold Start in Recommender Systems: A Systematic Literature Mapping
Authors:

Nathalia L. Cezar, Isabela Gasparini, Daniel Lichtnow, Gabriel M. Lunardi and José P. Moreira de Oliveira
Abstract: Recommender Systems are designed to provide personalized item recommendations to users based on their preferences and behavioral patterns, aiming to suggest items that align with their interests and profile. In Recommender Systems, a common issue arises when the user’s profile is not adequately characterized, particularly at the initial stages of using the system. This issue has persisted in Recommender Systems since its inception, commonly known as Cold Start. The Cold Start issue, which impacts new users, is called User Cold Start. Through a systematic literature mapping, this paper identifies strategies to minimize User Cold Start without reliance on external sources (such as social networks) or user demographic data for initializing the profile of new users. The systematic literature mapping results present strategies aimed at mitigating the User Cold Start Problem, serving as a foundational resource for further enhancements or novel proposals beyond those identified in the review. Thus, the goal of this work is to understand how to create an initial user profile before any prior interaction and without using external sources in the recommender system.
Download
Paper Nr: 73
Title:

Module of Contrastive Analysis for a Phonological Assessment Software in Development
Authors:

João B. Marques, João D. Lima, Márcia Keske-Soares and Fabrício A. Rubin
Abstract: The interest in software as tools to assist speech therapy has grown in recent years, with proposed features primarily focused on the analysis of children’s speech. However, there is still a gap in tools to apply phonological assessments that are suitable to collect data. In this context, our research group is working on a digital platform called “e-Fono”, which consists of a mobile application, a REST API, and a web service where the speech therapist has access to the assessments. However, despite the existence of robust tools for data analysis, predominantly used in academic contexts, no software was found with a module for phonological contrastive analysis in Brazilian Portuguese. In the contrastive analysis, the speech therapist compares and identifies the contrasts between the child’s speech and that of an adult, providing a detailed report on which phonemes and in which positions of the words the child experiences greater difficulties. Through this process, currently carried out manually by the speech therapists in our group, the child’s phonetic inventory is also obtained – a list of all phonemes the child can articulate. This paper proposes the development of a contrastive analysis module, which has been implemented in the e-Fono digital platform. In our implementation, the module was able to perform an automatic contrastive analysis by comparing the child’s phonetic transcriptions with known correct transcriptions from our database. The results can be reviewed by the speech therapist, who can replace and submit information of this analysis in case of wrong or incomplete results. With these information on the platform, it will be possible to identify speech difficulties in children and guide the speech therapist toward a specific treatment for them. Finally, in this paper we also present screens from the implemented prototype, which may be available to the general public after validation and adjustments with specialists from our group.
Download
Paper Nr: 128
Title:

Enhancement of the Online Presence of Small and Medium Sized Enterprises with Minimum Impact on Traditional Business Activities in Towns and Cities
Authors:

María Garrido, José J. Castro-Schez, Rubén Grande, Santiago Schez-Sobrino and David Vallejo
Abstract: This paper introduces an innovative strategy for an e-commerce portal designed to support small and medium-sized enterprises (SMEs), integrating local businesses not directly related to product sales, referred to as “satel-lite businesses”, such as bars, restaurants, cinemas or sports facilities. This proposal modifies the existing VR-ZOCO e-commerce portal structure to strengthen local economies, facilitating a symbiosis between online shopping and physical leisure activities. Following a purchase on the portal, users are offered the option to collect their products at specific local points. Linked to this collection act, personalized “leisure plans” are generated, based on “leisure activities” from satellite businesses. This initiative not only promotes the digital growth of SMEs but also encourages the revitalization and sustainable development of local communities. This paper details the fundamental concepts emphasizing how the interaction between online shopping and physical leisure activities can enrich the consumer experience and simultaneously support local businesses. The research proposes a balanced solution that aligns with modern consumer expectations and contributes to local economic and social growth, representing a significant advancement in the digital transformation of SMEs.
Download
Paper Nr: 25
Title:

Building Knowledge-Based Applications in Technical Guide Based on Semantic Web Technology
Authors:

Yu-Liang Chi and Han-Yu Sung
Abstract: In this study, a technical guide titled, “Prevention and Treatment of Pressure Ulcers/Injury: Clinical Practice Guidelines” (CPG) is transformed into a programming solution to provide knowledge-based queries. The three main development processes include the following: (1) Analysis of knowledge components from the CPG, which focuses on the manual extraction of programmable knowledge elements, structures, and relationships. (2) The development of knowledge models and corresponding instances, where Semantic Web technology, including the Web Ontology Language (OWL), is used to develop a knowledge model and resource description framework (RDF) to edit the CPG data as instances. (3) The development of knowledge query applications, which involves the development of a Web-based application system that includes a front-end user interface and backend applications. The tools used to develop this system include a server-side scripting language, SPARQL, and an RDF triplestore. Some advanced queries use semantic inference engines to provide implicit knowledge. Based on evaluation, the proposed system can facilitate users in obtaining more accurate guideline texts and related knowledge. Finally, the operating time of the system query can be reduced significantly compared with that of a manual query.
Paper Nr: 95
Title:

A Reflective Architecture for Agent-Based Models Applied to Social Network Sites
Authors:

Diego Nuñez, Tomás Vélez, Paul Leger and Daniel San Martín
Abstract: Social network sites serve as effective platforms for word-of-mouth marketing (WOM), often analyzed through Agent-Based Models (ABMs). However, implementing ABMs can be daunting, with programmers facing the choice of building from scratch or using frameworks. To tackle this, we propose FASOW (Flexible Agent Simulator for Open WOM) architecture, employing the Reflective Tower design. FASOW’s four layers cater to varying complexities, simplifying implementation by breaking down models into manageable sub-layers. We validate FASOW through a case study on Twitter, examining agent saturation effects in WOM marketing. Results indicate FASOW’s efficacy, though further use cases are needed for comprehensive evaluation. Additionally, we offer an online proof-of-concept for this architecture.
Download
Paper Nr: 165
Title:

Machine Learning Support for Time-Efficient Processing Dangerous Driving Detection Using Vehicle Inertial Data
Authors:

Matheus J. Silva de Almeida, Julia K. Ladeira, Caio G. Vicentin, Andre C. Costa, Marcia Pasin and Vinícius K. Marini
Abstract: Detection of dangerous driving behavior is a key component to improving road safety. It can be successfully carried out using data collected by sensors widely available in smartphones. Current work focuses on two groups: either they classify drivers in a binary way, into good and bad drivers, or they provide a scoring scale, allowing for a larger group of categories. This detection of dangerous driving behavior can be done with high granularity, evaluating a total distance covered by the driver on a trip, or with minute granularity, through the evaluation of small sections of driving, also making it possible to identify which maneuvers the driver is carrying out negligently. However, the process of collecting data for dangerous driving behavior is complicated because the driver needs to carry out these maneuvers, so that a classifier can later detect them, adding to situations of insecurity in traffic. Moreover, the solution needs to execute efficiently, so that the detection of dangerous driving behavior can be carried out in real time. Given this problem, we propose a time efficient dangerous driving detection system using vehicle inertial data. In contrast to other works, we collected data in a simulation environment with a model car that allows us to perform risky maneuvers, which would not be possible in a real environment. We identify in our small dataset the dangerous driving behavior pattern. Thus, given the established pattern, we applied a machine learning method to generate a classifier to enable the detection of dangerous driving behavior. The resulting system achieved a total average accuracy of 85.61% in our experiments using a small dataset as input towards efficient data processing.
Download
Paper Nr: 207
Title:

Designing and Building a Low-Cost IoT Solution for Natural Disaster Monitoring and Mitigation: An Experience Report
Authors:

Anselmo R. Costa, Claudia R. Motta and Daniel Schneider
Abstract: Mitigating natural disasters is a global concern and a social challenge, and new technologies and social inclusion mechanisms have been applied to improve prevention and response to these events. Several initiatives seek to mitigate natural disasters and rely on technologies such as the Internet of Things, crowdsourcing, volunteer training, society involvement, among others. Although there are many initiatives in this regard, building a model enabling real mitigation of natural disasters remains a major challenge. This study aims to present a Natural Disaster Mitigation model, which combines different technologies through the construction of a multidisciplinary group composed of researchers, students, civil defense technicians, and the municipal school network. The goal is to develop and implement a meteorological monitoring network using low-cost technological artifacts installed in school systems in the target cities.
Download

Area 5 - Human-Computer Interaction

Full Papers
Paper Nr: 34
Title:

Prediction of the Employee Turnover Intention Using Decision Trees
Authors:

Ana Živković, Dario Šebalj and Jelena Franjković
Abstract: This study examines the effectiveness of Decision Tree methodology in predicting employee turnover intention, an area in which this method has received limited research. In this paper, primary research was conducted and four Decision Tree algorithms were applied to a sample of 511 respondents. The study incorporates several predictor variables into the model, including job satisfaction, perceived organizational commitment, perceived organizational justice, perceived organizational support, and perceived alternative job opportunities, to assess their influence on turnover intention. The assessment measure of the model was Recall. The results indicate that the Decision Tree model using the RandomTree algorithm is relatively successful in predicting turnover intentions (almost 60% accuracy rate), with job satisfaction, especially opportunities for personal growth and affective organizational commitment being significant predictors. Other influencing factors include satisfaction with salary and the job itself, as well as interpersonal relationships. This study underscores the potential of the Decision Tree method in human resource management and provides a basis for future research on the role of predictive analytics in understanding employee turnover dynamics.
Download
Paper Nr: 49
Title:

Virtual Reality-Based Adapted Handball Serious Game for Upper Limb Rehabilitation in Spinal Cord Injured Patients
Authors:

J. Albusac, V. Herrera, Oscar Dominguez-Ocaña, E. Angulo, A. L. Reyes-Guzmán and D. Vallejo
Abstract: In this paper, we present a virtual reality-based serious game that simulates the training of a wheelchair handball goalkeeper. It is designed to complement traditional therapy for upper limb rehabilitation and trunk mobility improvement in spinal cord injury patients. The proposal is underpinned by a multi-layered architecture that provides a therapeutic environment that enhances patient motivation and satisfaction through gamification techniques. The architecture also provides precise kinematic recording during exercise performance, since the recorded data is essential for therapists to objectively assess each patient’s progress. Particularly, the recorded can be used to assess the extent of movement, how fast and smooth it is, the number of repetitions and their consistency, as well as the accuracy and precision of movements, balance, and posture control. The serious game was tested in the Hospital Nacional de Parapléjicos de Toledo, involving patients and healthcare professionals. The collected data are publicly available. This preliminary evaluation has been focused on assessing its functionality and safety. Following the exercise sessions, all participants were asked to complete a short questionnaire to measure their motivation, sense of achievement, satisfaction and overall comfort and well-being in the virtual environment. Future plans include expanding the patient sample and monitoring the long-term progress and impact of VR therapy on the recovery of mobility in the affected limbs.
Download
Paper Nr: 51
Title:

The Usability of Persistent and non-Persistent Headers on Web Pages
Authors:

Pietro Murano and Petter Kongshaug
Abstract: Websites are used for many different purposes and can also be designed in different ways with different styles. Certain website designs use persistent headers, while some alternatively use non-persistent headers. Online guidelines give ideas on how these are best used. However, there is no published systematic study investigating the performance and user satisfaction of these types of headers. In this paper we present an empirical experiment where persistent headers are compared with non-persistent headers. Two prototype news websites varied only in terms of their headers were developed and used in the experiment. The basic results suggest that persistent headers are more usable, particularly on a slightly larger screen. The analysis indicated that performance and user satisfaction are increased with the use of persistent headers on websites.
Download
Paper Nr: 89
Title:

Exploring Usability and User Experience Evaluation Methods: A Tertiary Study
Authors:

Geremias Corrêa, Roberto Pereira, Milene S. Silveira and Isabela Gasparini
Abstract: Usability and User Experience (UX) evaluation methods have important roles in business and scientific spheres, effectively pinpointing areas for enhancement across a broad spectrum of applications. Primary and secondary scientific studies investigating these methods are relevant and provide a panorama of different domains. While providing macro views on the topic is necessary, tertiary studies are still uncommon. This paper fills this gap by presenting a tertiary study conducted through a systematic search methodology, following Petersen’s guidelines. Studies indexed by Scopus, IEEE Xplore, and ACM search engines were considered, resulting in 487 retrieved studies, from which 36 were deemed relevant, and another 7 studies were added through a snowballing search strategy. From the selected studies, methods, domains of application, and considerations for the inclusion of accessibility in studies, among other information, were identified and discussed. Results revealed Questionnaires as the prevalent method in these studies, Brazil and Indonesia as the leading countries in authorship of publications, and Observation, Inspection, and Inquiry as the most common category for methods. These results suggest a prevalence of well-structured methods, generally with lower costs and application times, revealing space for further investigation.
Download
Paper Nr: 99
Title:

Enhancing Scientific Communication: Prioritizing User Experience in Audiovisual Interfaces
Authors:

Cintia B. Mesquita, Adriana B. Santos and Rogéria C. Gratão de Souza
Abstract: Audiovisual content has emerged as a powerful tool for scientific communication, enabling a broader reach, clearer explanations of complex topics, and greater recognition for researchers. To maximize its impact and fully unleash its potential, accessible and user-friendly interfaces are needed on platforms and mobile apps. This work prioritizes the identification of essential usability requirements for interfaces featuring scientific audiovisual content, aiming to streamline development efforts. Drawing from established literature and an analysis of existing web platforms and mobile apps, we have identified 25 key usability metrics to guide the creation of interfaces that prioritize user experience. Empirical evaluations of these platforms and apps have revealed significant deficiencies in meeting usability criteria such as effectiveness, efficiency, and user satisfaction.
Download
Paper Nr: 116
Title:

Advanced VR Calibration for Upper Limb Rehabilitation: Making Immersive Environments Accessible
Authors:

Vanesa Herrera, Ana Reyes-Guzmán, David Vallejo, José Castro-Schez, Dorothy N. Monekosso, González-Morcillo Carlos and Javier Albusac
Abstract: The creation of accessible spaces is essential for patients with motor injuries to conduct therapy safely and effectively. Disruptive technologies such as Virtual Reality (VR) are increasingly being used as a complement to traditional therapy, with excellent results. VR allows, among other things, the realistic recreation of physical spaces, so much so that it is relatively easy to run the risk of transferring physical barriers into the virtual space. This article proposes an innovative method of calibration in virtual environments that assesses the motor limitations of patients with cervical spinal cord injuries, doing so individually for each upper limb. The result is the dynamic adaptation of virtual environments to make them accessible and safe for rehabilitative therapy practices. This method has been integrated into the Rehab-Immersive platform, which hosts a series of serious games aimed at rehabilitating upper limbs, using immersive gamification techniques.
Download
Paper Nr: 170
Title:

Process and Challenges in Designing Data Registration via File Import in an Administrative e-Commerce System
Authors:

Pedro S. Silva, Anna Julia L. de Souza and Ingrid T. Monteiro
Abstract: This article presents the full account of a case study within VTEX, a Brazilian e-Commerce company. It reported the design challenge to conceive the functionality for importing records via a spreadsheet file. Several techniques of Human-Computer Interaction (HCI) and User Experience (UX) were applied within the Double Diamond design process, such as CSD matrix, ideation techniques, user stories, interaction diagrams, prototypes. In this document, we report how each technique was applied and present the results obtained from them. The results of the prototype usability test with user representatives are also presented. Some lessons learned, from this case study, were: to keep the design process iterative, exchanging techniques whenever necessary; to involve team members so that everyone is on the same page regarding the problem and the solution; to test the solution as soon as possible while still in prototype, even if it is not with end users but with their representatives.
Download
Paper Nr: 175
Title:

Validation and Refinement of Usability Heuristics for Interactive Web Maps
Authors:

Juliana O. Marquez, Paulo Meirelles and Tiago Silva da Silva
Abstract: The usability of interactive web mapping systems is crucial for a wide range of applications, extending from urban planning to personal navigation. This study, grounded in the context of Human-Computer Interaction (HCI), specifically aims to enhance the usability of interactive web maps. The goal is to provide designers and developers with improved guidelines that not only elevate the user experience but also effectively address the unique challenges of these interfaces, promoting more efficient navigation. Our methodology is distinguished by the development of new usability heuristics, derived from a detailed analysis of the specificities of interactive web mapping systems. The study proposes the introduction of a set of 12 usability heuristics, carefully adapted for these systems. The preliminary results are promising, outlining a set of heuristics that have the potential to be significant in the design and implementation of interactive web maps. The contribution of this study is substantial, offering new perspectives for the continuous improvement of usability heuristics and emphasizing the need for specific approaches for different digital interaction contexts. Thus, this work not only advances the theoretical field of HCI but also provides crucial practical guidelines for the future development of interactive web mapping systems, meeting the current demands and expectations of users.
Download
Paper Nr: 206
Title:

Understanding the Factors Influencing Self-Managed Enterprises of Crowdworkers: A Comprehensive Review
Authors:

Alexandre P. Uchoa and Daniel Schneider
Abstract: This paper investigates the shift in crowdsourcing towards self-managed enterprises of crowdworkers (SMECs), diverging from traditional platform-controlled models. It reviews the literature to understand the foundational aspects of this shift, focusing on identifying key factors that may explain the rise of SMECs, particularly concerning power dynamics and tensions between Online Labor Platforms (OLPs) and crowdworkers. The study aims to guide future research and inform policy and platform development, emphasizing the importance of fair labor practices in this evolving landscape.
Download
Paper Nr: 240
Title:

Operator Fatigue Detection via Analysis of Physiological Indicators Estimated Using Computer Vision
Authors:

Nikolay Shilov, Walaa Othman and Batol Hamoud
Abstract: The complexity of technical systems today causes an increased cognitive load on their operators. Taking into account that the cost of the operator’s error can be high, it is reasonable to dynamically monitor the operator to detect possible fatigue state. Application of computer vision technologies can be beneficial for this purpose since they do not require any interaction with the operator and use already existing equipment such as cameras. The goal of the presented research is to analyze the possibility to detect the fatigue based on the physiological indicators obtained using computer vision. The analysis includes finding correlations between the physiological indicators and the fatigue state as well as comparing different machine learning models to identify the most promising ones.
Download
Short Papers
Paper Nr: 47
Title:

Managing Adverse Commentary on Social Media: A Case Study of an Australian Health Organisation
Authors:

Gitte Galea, Ritesh Chugh and Lydia Mainey
Abstract: Health communication on social media is complicated, challenging, and multi-dimensional. Globally, the evolution of health communication has transformed rapidly from one-way to two-way interaction, with diverse audiences expressing limitless and often unconstrained commentary based on individual beliefs. This paper, a segment of a comprehensive doctoral study into the adoption and utilisation of social media within a large Australian health organisation, specifically Queensland Health, offers a snapshot of the research findings for managing negative commentary. This novel study interviewed social media administrators to understand their experiences and perceptions of social media use, underscoring the prominence of negative commentary as a notable drawback to the effective use of social media. Paradoxically, such adverse commentary also catalyses discussions and leads to helpful feedback. Effectively managing unacceptable commentary necessitates the implementation of a strategic response complemented by adequate resources and training.
Download
Paper Nr: 67
Title:

Usability and User Experience Questionnaire Evaluation and Evolution for Touchable Holography
Authors:

Thiago D. Campos, Eduardo F. Damasceno and Natasha C. Valentim
Abstract: Augmented and Mixed Reality enables applications where users engage in natural hand interactions, simulating touch, termed touchable holography solutions (THS). These applications are achievable through head-mounted displays and are helpful in training, equipment control, and entertainment. Usability and User eXperience (UX) evaluations are crucial for ensuring the quality and appropriateness of THS, yet many are assessed using non-specific technologies. The UUXE-ToH questionnaire was proposed and subjected to expert study for content and face validity to address this gap. This study enhances questionnaire credibility and acceptance by identifying clarity issues, aligning questions with study theory, and reducing author-induced bias, offering an effective and cost-efficient approach. The study garnered numerous contributions that were analyzed qualitatively and processed to refine the questionnaire. This paper introduces the UUXE-ToH in its initial version, details expert feedback analysis, outlines the methodology for incorporating suggestions, and presents the enhanced version, UUXE-ToH v2. This evidence-based process contributes to a better understanding of usability and UX evaluation in the THS context. UUXE-ToH can impact the quality of life of users of solutions applied to education, health, and entertainment by helping develop better products.
Download
Paper Nr: 68
Title:

Supporting User-Centered Requirements Elicitation from Lean Personas: A UX Data Visualization-Based Approach
Authors:

Maylon Macedo, Gabriel V. Teixeira, Ariel S. Campos and Luciana Zaina
Abstract: The interest in exploring User Experience (UX) data to support the requirement elicitation of interactive systems is not new. Although the literature discusses traditional methods to gather UX data (e.g., interviews and surveys), personas have arisen as a more user-centered technique that presents information about users’ needs, preferences, and characteristics about the application domain. Nonetheless, persona data is often represented from qualitative data in textual format, which can insert difficulties in browsing and exploring a set of personas. This paper aims to present visualizations designed using Information Visualization (InfoVis) principles that support the navigation and search-out UX data in a persona dataset. To aid in eliciting UX-related requirements, our visualizations are based on a funnel perspective that guides designers and developers to examine data from an overview first, then zoom and filter, to achieve the qualitative data in detail finally. We evaluated with 20 participants concerning the interpretation of the visualizations. The results revealed that the participants, even those with little experience in requirement elicitation, could interpret and find relevant UX data from the visualizations.
Download
Paper Nr: 80
Title:

Guiding the Adoption of UX Research Practices: An Approach to Support Software Professionals
Authors:

Maria Anita de Moura, Suéllen Martinelli and Luciana Zaina
Abstract: The interest in User Experience (UX) with interactive products and services has grown in the industry. In this context, the research with end-users contributes to articulating practices, methods, and research techniques on UX that can be applied at different stages of software development. Nevertheless, software development professionals have demanded tools that can aid them in selecting the suitable method or technique for a given purpose of user research. To address this demand, we developed guidelines that suggest methods and techniques for working with user experience research. Considering the guidelines, we created the GURiP tool in the virtual catalog format, providing a more dynamic interaction with the guidelines. We evaluated the proposal acceptance with 32 software professionals from software startups and established companies. Our results revealed that professionals of both types of companies showed similar acceptance and reported more positive than negative feedback about the guidelines. We also found that participants’ profiles, such as years of experience or affiliation with startups or established companies, did not influence the acceptance of the guidelines.
Download
Paper Nr: 102
Title:

Challenges in Metaverse Adoption on People, Process, and Technology Perspectives: A Review from the Five past Years
Authors:

Simone C. dos Santos, Matheus B. Nobre, Sidarta L. Varela, Felipe Figueroa and Dairon E. Martins
Abstract: The Metaverse represents a virtual world intrinsically connected to reality. Its essence lies in constructing a digital space encompassing different media types, merging uniquely with the real world. This environment’s innovation and its potential users’ attraction to immersive experiences bring countless opportunities for individuals and organizations, as highlighted in the Gartner Group Report on the "Top Strategic Technology Trends for 2023." However, significant opportunities also bring significant challenges. So, in this context, this paper presents the results of a Systematic Literature Review (SLR) study focusing on the challenges of using the Metaverse from the perspective of people, processes, and technology. Following Kitchenham and Charters (2007) guidelines, we analysed 55 studies from relevant sources to understand the intricate interaction between human elements, procedures involved, and technology within the context of the Metaverse. This review made it possible to present a comprehensive view of the challenges and obstacles in this field of investigation, offering insights into the quantity and quality of available evidence. The challenges identified in this study summarize the main academic contributions related to using the Metaverse in the last five years.
Download
Paper Nr: 107
Title:

Crafting a Journey into the past with a Tangible Timeline Game: Net Als Toen as a Tool to Enhance Reminiscence in Elderly with Alzheimer's Disease
Authors:

Renske Mulder, Puck Kemper, Hannah Ottenschot and Anis A. Hashim
Abstract: Alzheimer’s Disease (AD) poses significant challenges for individuals and their caregivers due to its impact on memory, behavior, and cognitive abilities. With the projected increase in AD cases in the coming years, innovative technologies are needed to address the growing demand for elderly care and support for people with AD. Reminiscence therapy (RT) can have positive effects on the rate at which AD symptoms worsen. This paper presents an interactive game based on RT called Net Als Toen, which serves as a conversation starter. The ideation phase, lo-fi prototype development, and hi-fi prototype testing are discussed. Results from playtests show that the embedded reminiscence theory in Net Als Toen can help people with AD in talking about their memories. Additionally, results suggest that personalization options and improved user interface elements are important in making the application successful. Overall, this paper contributes to developing a social game based on RT, focusing on interpersonal reminiscence therapy, to foster interactive conversations and enhance the well-being of individuals with AD.
Download
Paper Nr: 120
Title:

Exploring Interaction Mechanisms and Perceived Realism in Different Virtual Reality Shopping Setups
Authors:

Rubén Grande, Javier Albusac, Santiago Sánchez-Sobrino, David Vallejo, José J. Castro-Schez and Carlos González
Abstract: Within the e-commerce field, disruptive technologies such as Virtual Reality (VR) are beginning to be used more frequently to explore new forms of human-computer interaction in the field and enhance the shopping experience for users. Key to this are the increasingly accurate hands-free interaction mechanisms that the user can employ to interact with virtual products and the environment. This study presents an experiment with a set of participants that will address: (1) users’ evaluation of a set of pre-formalised interaction mechanisms, (2) preference for a large-scale or small-scale shopping environment and how the degree of usability while navigating the large-scale one, and (3) the usefulness of monitoring user activity to infer user preferences. The results provided show that i) interaction mechanisms made with users’ hands are fluid and natural, ii) high usability in small and large shopping spaces and the second ones being preferred by the users and iii) the recorded interactions can be employed for user profiling that improves future shopping experience.
Download
Paper Nr: 137
Title:

Evaluating the Acceptance and Quality of a Usability and UX Evaluation Technology Created for the Multi-Touch Context
Authors:

Guilherme K. Filho, Guilherme C. Guerino and Natasha C. Valentim
Abstract: The multi-touch context is a poorly explored field when it comes to usability and User eXperience (UX) evaluation. As any kind of system, it must be properly evaluated in order to be truly useful. A Systematic Mapping Study (SMS) showed that there is no technologies being used to evaluate the UX and usability of multi-touch systems that were specifically built for it. The use of generic technologies can leave behind important perceptions about the multi-touch systems specificites. To fill this gap, the User eXperience and Usability Multi-tiuch Evaluation Questionnaire (UXUMEQ) was created. UXUMEQ is a questionnaire that seeks to evaluate the UX and usability of multi-touch systems taking into account the most relevant aspects being used to this end, such as performance, workload, intuition, error tolerance and others. As any new technology, UXUMEQ must be evaluated in order to be improved. In this paper, we carried out a quantitative analysis to verify the public acceptance of UXUMEQ when compared with generic technologies being used to evaluated multi-touch systems. This analysis showed greater public acceptance about UXUMEQ regarding usefulness and ease of use. We also invited Human-Computer Interaction (HCI) experts to inspect UXUMEQ through a qualitative study. Their perceptions were collected and evaluated through the Grounded Theory method, that will contribute to provide a most refined version of UXUMEQ.
Download
Paper Nr: 157
Title:

The Influences of Employees' Emotions on Their Cyber Security Protection Motivation Behaviour: A Theoretical Framework
Authors:

Abdulelah Alshammari, Vladlena Benson and Luciano Batista
Abstract: At the employee level, cyber threats are a sensitive issue that requires further understanding. Cyber-attacks can have a multifaceted impact on an organisation. Psychological research has demonstrated that emotions influence individuals’ motivation to engage in cybersecurity protection behaviour. Most extant research focuses on how external influences may affect employees’ cyber security behaviours (e.g., understanding risk, rationality in policy decision-making, security regulations, compliance, ethical behaviour, etc.). Little research has been done to date on how employees’ internal emotions affect their motivations for cybersecurity protection. To bridge this gap, this paper aims to expand the research by establishing a model for measuring the effect of employees’ negative and positive emotions on their cybersecurity protection motivation behaviour. The model will emphasise self-efficacy as a mediating factor and cybersecurity awareness as a moderating factor, and this model is based on a comprehensive evaluation of the existing literature. More specifically, the proposed theoretical model was established by integrating the protection motivation theory (PMT), and the broaden and build theory (BBT) to understand the effects of negative and positive emotions on employees’ cybersecurity protection motivation behaviour. This study opens the gates for future research on the role of emotions on employees’ cybersecurity protection motivation behaviour. Furthermore, understanding how emotions affect employees’ cybersecurity protection motivation will be a valuable contribution to academia, helping decision-makers and professionals deal with the effects of emotions regarding cybersecurity.
Download
Paper Nr: 40
Title:

Design Analysis of Smart Water Meters: An Open Design Approach
Authors:

Flávio H. Alves, Maria C. Baranauskas and Alexandre L’Erario
Abstract: The control of water resources has become increasingly necessary due to population growth and climate questions. This control is so critical that it is related to the UN’s Sustainable Development Goals (SDGs): SDG #6 is clean water and sanitation, which aims to ensure the availability and sustainable management of water and sanitation for all people. This study addresses the application of Information and Communication Technologies (ICT) as a solution to water management problems in smart cities, highlighting the importance of the participatory construction of smart water meters (SWMs) as a potential solution to monitor water consumption. The design and implementation of SWM provides significant technical and social challenges, but promises to improve efficiency in water control and consumption. The objective of this study is to investigate SWMs, analyzed using the open design methodology to identify problems, questions, solutions and ideas in their implementation. We understand how SWMs can benefit water management and propose more effective and participatory solutions. The methodology adopted involves three stages: analysis of the case of Company X, selection of cases from the literature, and analysis with Open Design. Analysis is performed using Open Design artifacts to identify stakeholders, questions and problems, and solutions and ideas related to SWMs. As a result, it is possible to identify open problems and questions that need to consider a more participatory and inclusive approach in developing SWM solutions. The use of Open Design is promising and makes stakeholder engagement more effective in creating sustainable and affordable solutions.
Download
Paper Nr: 251
Title:

Industry 4.0: Wearable IoT Device Applied to Warehouse Management
Authors:

Lucas S. Vicente, Saul Delabrida, Mateus C. Silva, Adrielle C. Santana and Ricardo R. Oliveira
Abstract: Companies in the retail sector need proper control of their stock to avoid financial waste and guarantee the effectiveness of their operations. A more detailed analysis of this problem reveals the complexity of implementing a management methodology that enables optimal control of all stock, since human errors occur during operations and various scenarios depend on different variables. Therefore, to solve the problem of efficient warehouse stock management and the resulting inefficiency of operations, this study proposes the implementation of a wearable, developed using a Raspberry Pi 4B with IoT and Node-Red, in conjunction with a mobile device, which assists operators during the processes of stocking, searching for and removing material from the warehouse more efficiently. As a result, the proposed system can identify, by reading an RFID tag with a mobile device, the characteristics of the equipment in question, showing all this information on an OLED display, as well as directing what will be done with this equipment via an app. Among the metrics that demonstrate the effectiveness of the proposed system is the time taken to stock and remove the material, since all the procedures are managed in real-time on the app and updated in its inventory control.
Download

Area 6 - Enterprise Architecture

Full Papers
Paper Nr: 41
Title:

Conformance Checking on Timed Automaton Process Models
Authors:

Sohei Ito and Kento Hamae
Abstract: Conformance checking is a kind of process analysis methods that evaluate the difference between modeled behavior and recorded behavior of a process. Usual conformance checking evaluates such difference (called fitness ) based on only the order of executed tasks. Therefore, one cannot correctly evaluate fitness for process executions that violate the task completion time required by the model. In this paper, we consider process models with time constraints given as timed automata, and propose a method for evaluating fitness of timed traces with timed automaton process models. We implement two algorithms to compute the fitness we proposed, namely a naive exhaustive search and A* algorithm, and show experimental results with simple process models.
Download
Paper Nr: 84
Title:

Towards a Link Mapping and Evaluation Approach for Core Operational Business-IT Alignment
Authors:

Ali Benjilany, Pascal André, Dalila Tamzalit and Hugo Bruneliere
Abstract: Business-IT Alignment (BITA) is an important mean of evaluating the performance of IT systems operating within a business organisation. The software architects’ need for representing, analysing and interpreting the alignment situations remains among the main challenges. Despite different initiatives in the last two decades, available solutions remain too diverse and limited. As a consequence, more methodological guidelines are still needed to improve the support for BITA. In this paper, we address Core Operational BITA (COBITA) as a subset of BITA which targets the operational integration of business and application artefacts. To this end, we first propose two types of COBITA links between the business layer and the application one. We propose then an approach for establishing these links and evaluating them. The objective is to provide indicators for domain experts and software architects to assess the quality of the alignement between the two layers. We decided to choose Archimate, a standard language, to model the business and application layers. Then, we specify the two types of COBITA links to establish a mapping between the business and applications layers. Finally, we rely on the obtained cartography to evaluate the alignment via a set of proposed metrics and consistency rules. An initial version of the approach has been implemented in the Archi tool, and we experimented with it on the SoftSlate system.
Download
Paper Nr: 126
Title:

Validating a Practical Method for Planning Co-Evolution of Business and IT in a Public Sector Organisation
Authors:

Christof Meekers, Sara Nodehi, Tim Huygh, Laury Bollen and Joost Visser
Abstract: Background Business-IT alignment (BITA) remains a top management concern. A method for business-driven planning of changes in information systems was proposed by Nodehi et al., and evaluated in a limited way through an educational pilot and expert interviews. Aim We want to evaluate the proposed method in the public sector and identify possible improvements. Method We conducted a case study, involving four evolution projects in a local government organisation. We identified BITA challenges in the case organisation, then moderated the application of the method by participants, and finally studied the usefulness of the method through surveys and interviews. Results We identified 14 categories of BITA challenges in our public sector case organisation that largely match the challenges identified by Nodeehi et al. in the private sector. Overall, the method was well-accepted and highly appreciated by all participants. Several improvement points were identified. Clear links were found between identified BITA challenges and specific elements of the method. Conclusion The proposed planning method was found to be beneficial for improving BITA in the public sector. Additionally, the links found between BITA challenges and method elements provide insight in how the method helps achieve alignment.
Download
Paper Nr: 138
Title:

Modernization of Legacy Systems to Microservice Architecture: A Tertiary Study
Authors:

Nathalia D. Almeida, Gabriel N. Campos, Fernando D. Moraes and Frank J. Affonso
Abstract: Modernization of legacy systems to Microservice Architecture (MSA) has been a subject of interest in both academia and industry. MSA facilitates the development of systems by composing them as a collection of small, loosely coupled, and independent services. Despite the importance of this architectural style for system modernization, there is a lack of comprehensive study of how the modernization process must be conducted, regarding the main steps that must be executed, besides architectural patterns that may be used during this process. The goal of this paper is to provide a panorama and analysis, providing a panorama on the modernization of legacy systems to MSA. To do so, we conducted a tertiary study following the guidelines of a Systematic Mapping Study, based on 20 secondary studies. Our overview reveals some evidence related to modernization, such as the main motivations, the main architectural patterns, and the macro steps of a process. Moreover, a discussion of the main findings is addressed in our study. As the study presented in this paper addresses a research theme that is constantly evolving, the presented panorama may be an important guide for researchers and practitioners in the development of future work to advance and consolidate such theme.
Download
Short Papers
Paper Nr: 11
Title:

RAMOM: A Reference Architecture for Manufacturing Operations Management Activities in Industry 4.0
Authors:

Gonçalo Freire and André Vasconcelos
Abstract: Industry 4.0 has revolutionized manufacturing by introducing technologies such as Cyber-Physical Systems, Internet of Things and others that make manufacturing more efficient and dynamic. Despite these benefits, Industry 4.0 has a high barrier to entry. The complexity of manufacturing systems will inevitably increase, and it is also necessary to redesign existing manufacturing processes to take advantage of Industry 4.0. In this paper we use Enterprise Architecture to help companies to deal with the increasing complexity when adopting Industry 4.0. In our research, we found that many solutions have been developed to help companies make the technological transition to Industry 4.0, but none helps companies align their newly acquired technological capabilities with their production processes. To address this gap, we developed RAMOM, a reference architecture for manufacturing operation management activities in Industry 4.0. RAMOM is composed of several views, developed in the Archimate language, that provide information on the actors, functions, data types and how these relate to manufacturing operation management activities, thus guiding organizations in their implementation. To confirm its validity, we conducted an evaluation of RAMOM based on expert knowledge and an application of RAMOM in a Portuguese industry case study. We concluded that is useful to use RAMOM to help organizations adapt their processes to Industry 4.0.
Download
Paper Nr: 18
Title:

Enterprise Architecture Governance of Excellence
Authors:

Peter Hillmann, Mario Kesseler, Diana Schnell, Goran Mihelcic and Andreas Karcher
Abstract: Every organization has an architecture that must be adapted to the given circumstances and future needs in order to stay competitive. Particularly in the case of federated structures, there is a high degree of complexity in terms of both business and IT. For a smooth process and a goal-oriented management, this should be done as proactively as possible in an orderly manner. To make this possible, the principles of governance have to be applied. This forms the organizational basis for Enterprise Architecture Management. It creates high-quality information that is used for strategic and operational decisions. The challenge we address is the lack of Enterprise Architecture Governance with focus on a federated environment. Our goal is a detailed and applicable concept for Enterprise Architecture Governance. For this purpose, the several components are examined more closely and detailed explanations are given in a compact form. The evaluation is based on practical examples in both industry and government.
Download
Paper Nr: 20
Title:

An Enterprise Architecture Approach to Semantic Blockchain Interoperability
Authors:

Sebastião Sotto-Mayor, Rafael Belchior, Miguel Correia and André Vasconcelos
Abstract: Blockchain technology has revolutionized the way data is stored and accessed in a decentralized manner. However, the lack of interoperability between such systems is an ongoing challenge hindering their wider adoption. This document proposes a two-part solution composed of activities that aim to enhance semantic interoperability between homogeneous and heterogeneous blockchain systems. The first part are the design-time activities that consist of constructing an Archimate model, extracting its Resource Description Framework (RDF) ontology, and assessing its correctness utilizing a semantic reasoner. The second part are the runtime activities that involve leveraging the resulting ontology in a supply chain management application to validate transactions among participants in a network of systems. The evaluation results are promising, demonstrating that a shared ontology can support a transparent and accurate transaction validation approach. Thus, this work is a significant step in proving that distributed ledger technologies can benefit from enterprise architecture techniques to improve their interoperability.
Download
Paper Nr: 22
Title:

An Enterprise Architecture-Based Approach Towards More Agile and Resilient Command and Control
Authors:

Ovidiu Noran and Peter Bernus
Abstract: Challenges to the global power balance leading to the possibility of peer warfare and technological advances enabling autonomous and swarming-capable vehicles performing increasingly complex operations in contested environments have prompted a stringent need towards more agile and resilient Defence doctrines. On the other hand, Disaster Response efforts also require a more versatile and robust Command and Control (C2) approach in the context of increasing intensity and frequency of natural disasters triggered by climate change. The efforts towards addressing these C2 challenges typically consider the required aspects in isolation, although more often than not they are closely related and as such, changes to one C2 aspect may have unintended effects on the others. Therefore, a holistic approach is required considering the overall effects of the envisaged transformation, so as to maintain the consistency of the C2 evolution effort. This paper proposes such an integrated method that employs Enterprise Architecture modelling artefacts facilitating an overarching approach towards more agile and resilient C2 evolution. A case study is employed to illustrate the concepts proposed and further analyse the relation between the changed warfare and disaster response paradigms and more agile and resilient C2 approaches.
Download
Paper Nr: 32
Title:

An Architectural Viewpoint for Managing BizDevOps Software Projects
Authors:

Guillermo Fuentes-Quijada, Francisco Ruiz-González and Angélica Caro
Abstract: BizDevOps extends the DevOps approach by integrating a business cycle that encompasses stakeholders beyond the realm of information technology, aiming to support the alignment between IT and business while also fulfilling organizational objectives. This approach could be augmented with the use of enterprise architecture descriptions and customized architectural viewpoints, which facilitate the analysis, communication, and management of team-specific concerns. This study introduces BizDevOps-VP, proposing a viewpoint designed to enhance communication and decision-making within BizDevOps teams, focusing on their concerns, with the goal of supporting IT/business alignment without compromising agility.
Download
Paper Nr: 55
Title:

Speeding Up the Simulation Animals Diseases Spread: A Study Case on R and Python Performance in PDSA-RS Platform
Authors:

Rodrigo Schneider, Felipe Machado, Celio Trois, Glênio Descovi, Vinícius Maran and Alencar Machado
Abstract: The control and prevention of livestock diseases play a crucial role in safeguarding business continuity, simulating disease prevention and control measures are vital to mitigate future epidemics. In this sense, modelling systems can be an effective tool that allows the simulation of different ways of spreading diseases by configuring parameters allowing testing of different prevention measures. This work investigates enhancing a system that simulates disease spread processes in animals. The stochastic model system was developed in R; however, given a large amount of data and intense processing of stochastic functions that simulate spreading and control actions, it required optimization. We focused on translating and modifying it to Python using packages focused on data analysis, aiming to speed up the system execution time. We conducted experiments comparing high computational cost functions executed in the actual model R with the new proposal implemented in Python. The results showed that rewriting the code in Python has advantages such as performance in time execution, which in Python is more than four times faster than R, memory usage consumption in R uses 460 MB and 315 MB in Python.
Download
Paper Nr: 173
Title:

Job Crafters Going Digital: A Framework for IT-Based Workplace Adaption
Authors:

Angelina C. Schmidt, Michael Fellmann and Jakob Voigt
Abstract: The changing world of work is leading more and more people to reflect on the meaning and organization of their work. Increased flexibility allows individuals to define and shape their own jobs. However, adapting one’s job, which is referred to as job crafting, is a challenging manual task since many variables can be modified with unclear dependencies. Hence, to systematically promote job crafting behaviors, Job Crafting Information Systems (JCIS) were proposed a decade ago. However, up to now, it is highly unclear which IT-supported interventions could be implemented in such systems. Against this gap, we develop an integrated model that matches the different job crafting behaviors discussed in the literature with supporting and facilitating IT components. As a result of our literature review, we include the functional IT components recommendation, coaching, time management, and complaint management and identify gamification, simplification, prediction, and integration as important non-functional characteristics of JCIS.
Download
Paper Nr: 193
Title:

Business Process Improvements in Hierarchical Organizations: A Case Study Focusing on Collaboration and Creativity
Authors:

Simone D. Santos, Malu Xavier and Carla Ribeiro
Abstract: This paper describes a case study of business process improvement (BPI) in a large and hierarchical organization in the public sector. Business Process Management (BPM) is crucial in the inevitable digital transformation of large organizations, streamlining workflows and enhancing efficiency. It involves systematic design, execution, and continuous improvement of processes, incorporating efficient activities and digital tools like automation and artificial intelligence. Despite the benefits, implementing BPM in highly hierarchical organizations poses challenges, including resistance to change and communication barriers. Thus, the paper advocates a collaborative and creative BPI approach to address these as a crucial stage of the BPM cycle. Collaboration is essential for breaking down silos and promoting a holistic BPM approach, while creativity facilitates transformative change in established norms. From several BPI methodologies available, we select and apply one called Boomerang in a collaborative workshop format. This methodology is based on a design thinking process and gamification strategy. A case study utilizing Boomerang demonstrates successful BPI by balancing established structures with innovative transformations. Still, lessons learned are identified, emphasizing the need for careful preparation of a collaborative workshop, stakeholders’ selection, a kit of artifacts to support this event, and a trained group to conduct the BPI process.
Download
Paper Nr: 30
Title:

A Formal Execution Semantics for Sophisticated Dynamic Jumps Within Business Processes
Authors:

Thomas Bauer
Abstract: At business processes (BP) execution, in exceptional cases (e.g. to save time or to correct errors), users must have the possibility to jump forward and backward in the BP. Currently, this topic is hardly respected in scientific literature and only insufficiently realized by commercial BP engines. This paper develops a formal execution semantics for dynamic jumps. It does not only respect simple forward and backward jumps within sequences of activities, but comprehensive requirements as jumps into and out of parallel branches or within loops. Furthermore, the intended behaviour of concerned activities can be modelled, i.e., they may be caught up later or their results (output data) may be preserved and reused at their later repeated execution after a backward jump.
Download
Paper Nr: 57
Title:

Carbon-Aware Process Execution for Green Business Process Management
Authors:

Philipp Hehnle, Maximilian Behrendt, Luc Weinbrecht and Carl Corea
Abstract: Traditional business process management (BPM) focuses on the improvement of performance dimensions such as time, costs, and quality. Ecological aspects are usually not considered as an equal performance dimension. In this context, Green BPM approaches have been proposed to strengthen the awareness among people and organisations about the impact of business processes on the climate. However, little research in Green BPM covers the runtime of digitised processes, or provides concrete means to reduce carbon emissions during process execution. Therefore, we present an approach for carbon-aware process execution, which allows to automatically postpone energy-intensive activities to times when energy with low CO 2 emissions, e.g. solar energy, is better available. Importantly, our approach considers and complies with external regulations such as Service Level Agreements (SLAs) when postponing activities. Our approach is implemented in Camunda and has been evaluated in interviews with domain experts.
Download
Paper Nr: 78
Title:

The Role of Digital Artifacts in Fostering Ecosystem Creation
Authors:

Edoardo Meraviglia, Jacopo Manotti, Davide Moiana, Antonio Ghezzi and Andrea Rangone
Abstract: Since 2008, blockchain has slowly entered our lives in different ways: from Bitcoins and crypto currencies to platforms like Ethereum, to NFTs and Fan Tokens. There is often still speculation about these issues, but there are successful use cases that have revitalized traditional industries: this is the case in the sports industry. Many sports teams, particularly football teams, have always tried to monetize their fans in various ways to create new sources of revenue; and thanks to Fan Tokens this has been possible. Consequently, an exploratory multiple case studies was conducted in order to analyze how a digital artifact (Fan Token) can create an ecosystem that helps sports companies to innovate and monetize through increased fan engagement. As a final result, this research has led to the definition of an empirical model that demonstrates how the intrinsic characteristics of a digital artifact can be harnessed to create an ecosystem to create and capture value from fan-customers.
Download
Paper Nr: 92
Title:

AI Technology Adoption & Sustainability Improvement Though Cloud Solutions
Authors:

Maarten Voorneveld
Abstract: Cloud and AI are game changers in digital transformation, as it facilitates long-term digital development and technology adoption. A study of over 1000 organizations in Western Europe was conducted to identify company adoption of AI technology and cloud computing-based sustainability benefits. This paper offers the survey results and situates them within the larger context, showing how businesses employ cloud technology to achieve their AI and sustainability goals. Digital innovations such as AI technology are being realized via cloud services, allowing companies to better develop their product and services.
Download
Paper Nr: 94
Title:

Understanding the Interplay Between Startups and Accelerators for Early-Stage Resource Mobilization
Authors:

Davide Moiana, Jacopo Manotti, Antonio Ghezzi and Andrea Rangone
Abstract: Startups, representing the engine of innovation and technology entrepreneurship, face the challenge of securing resources for sustainable growth while generating innovative solutions. Startup accelerators have rapidly emerged as prominent players in the entrepreneurial ecosystem, providing resources, mentorship and training to startups. However, a deeper analysis of how startups approach accelerator programmes is often overlooked in the literature. Drawing on a multiple case study of 9 AI-based startups located in Italy that participated in different acceleration programmes, we explore how startups’ teams engage with acceleration programs. We find that early-stage startups engage with accelerators that focus on learning and validation mechanisms with the aim of searching for and accessing human capital, while they turn to accelerators that focus on access and reach mechanisms with the aim of pursuing market access and scaling objectives. The implication of these research could benefit both theory and practice by enhancing the understanding of the interplay between startups and accelerator programs, and by offering insights to founders to align participation with the stage and goals of their startups.
Download
Paper Nr: 140
Title:

Enabling Sustainability Due Diligence in Value Chains Through DLT-Based Governance
Authors:

Jan Zedel and Felix Eppler
Abstract: The European Union’s Corporate Sustainability Due Diligence Directive (CSDDD) represents a transformative approach to ensuring sustainable and ethical supply chain practices. However, the directive introduces complexities and significant efforts in the compliance processes of affected companies. This paper explores the question of how decentralized resource pooling can lead to more efficiency and lower costs in the compliance processes of the CSDDD. We propose a novel solution leveraging distributed ledger technology (DLT), specifically a token-curated registry (TCR) solution augmented with shielded transactions to ensure participant anonymity and trust. Our proposed approach could reduce redundancies, increase transparency, and facilitate a more robust risk analysis over the whole value chain. Our paper follows a design science research approach where we (1) identify the challenges posed by companies that fall under the new CSDDD regulation framework, (2) define the objectives of a distributed ledger-based solution and (3) design and develop a technical framework based on a TCR to handle parts of the CSDDD in an efficient and cost-effective way, setting a precedent for future empirical studies and practical implementations in the field of distributed ledger-based information systems in the field of sustainability corporate governance.
Download
Paper Nr: 187
Title:

Six Board Roles for Information Security Governance
Authors:

Sara Nodehi, Tim Huygh and Laury Bollen
Abstract: As cyber threats evolve, board engagement is becoming increasingly essential to ensure Information Security (InfoSec) is integrated into an organization’s strategic fabric, ensuring the protection of business value. Only through board-level active participation can the organization develop a security-conscious culture. Ultimately, board commitment to InfoSec helps reduce risks, maintain stakeholder trust, and ensure long-term success. However, little is yet known about the board’s exact role in Infosec. Leveraging a framework from corporate governance literature identifying board roles, and drawing parallels with extant InfoSec literature, this paper explores board-level involvement in InfoSec in greater depth, leading to the identification and description of the board of directors’ roles in this context. Moreover, the paper identifies a future research agenda to be pursued in an empirical setting to contribute to the growth of knowledge regarding board-level InfoSec governance.
Download
Paper Nr: 195
Title:

Redefining Data Governance: Insights from the French University System
Authors:

Guy Melançon, Nathalie Pinède and Ugo Verdi
Abstract: This position paper highlights the distinctive features of French universities that render data governance within these institutions a particularly challenging endeavor. These universities inherently operate in an exceptionally open milieu, necessitating the conceptualization of governance as a dynamic and adaptable framework that converges seamlessly with the governance structures of other institutions. The principle of collegiality further mandates a distributed approach to data governance, encompassing responsibilities, rules, and procedures across various levels of management. Moreover, it is essential to reevaluate the conventional viewpoint that segregates administrative tasks from research and teaching functions. Our findings underscore the necessity for developing and executing a dynamic, multi-tiered data governance model that integrates the three fundamental missions of universities. Given the intrinsic nature of French universities, it is imperative to envisage governance as an evolving ecosystem of agents who assume complementary responsibilities in a harmonized manner.
Download
Paper Nr: 254
Title:

An Architecture Framework for Higher Education
Authors:

Siegfried Rouvrais and Sobah A. Petersen
Abstract: In the realm of higher education, an educational architecture framework can play a pivotal role in fostering enhanced communication between program leaders and various educational stakeholders. Within this context, architecture views serve as comprehensive representations of the overarching architectural landscape, catering to the diverse requirements and needs of involved stakeholders. Embracing a view-based approach empowers higher education institutions to reinforce strategic alignment while seamlessly integrating change management practices to accommodate evolving requirements. In this perspective, this paper proposes six distinct views to reflect on how enterprise architecture could be applied to higher education. Examples are given based on ArchiMate models. These examples serve as compelling illustrations of how educational architecture frameworks can drive organizational transformation.
Download