ICSBT 2024 Abstracts

Area 1 - Business Intelligence

Full Papers

Paper Nr:	12
Title:	Comparative Analysis of Machine Learning Techniques for DDoS Intrusion Detection in IoT Environments
Authors:	Godwin Chukwukelu, Aniekan Essien, Adewale I. Salami and Esther Utuk
Abstract:	This study addresses the challenge of Distributed Denial of Service (DDoS) attacks in the Internet of Things (IoT) environment by evaluating the effectiveness of Intrusion Detection Systems (IDS) using machine learning techniques. Due to the lightweight computational configuration of IoT systems, there is a need for a classifier that can efficiently distinguish between legitimate and malicious network traffic without demanding substantial computational resources. This research presents a comparative analysis of four machine learning models: (i) k-Nearest Neighbour (k-NN), (ii) Support Vector Machine (SVM), (iii) Random Forest (RF), and (iv) Multilayer Perceptron (MLP), to propose a lightweight DDoS intrusion detection classifier. A novel classification model based on the MLP architecture is proposed, focusing on minimalistic design and feature reduction to achieve accurate and efficient classification. The model is tested using the CICIDS2017 dataset and demonstrates high accuracy and computational efficiency, making it a viable solution for IoT environments where computational resources are limited. The findings show that the proposed µML-IDS model achieves an accuracy of 99.8%, F-score of 96.5%, and precision of 99.96%, with minimal computational overhead, highlighting its potential for real-world application in protecting IoT networks against DDoS attacks.
Download

Paper Nr:	24
Title:	Sentiment Analysis-Based Chatbot System to Enhance Customer Satisfaction in Technical Support Complaints Service for Telecommunications Companies
Authors:	Anghelo Juipa, Luis Guzman and Edgar Diaz
Abstract:	In the competitive world of telecommunications, a good customer technical support complaint service can make a difference. However, this business process still presents deficiencies in its quality. In the capital of Peru, there were 102,665 internet complaints and 38,621 cable television complaints. 9.27% and 9.97% of these, respectively, weren’t resolved. In this sense, this research proposes the implementation of a chatbot, which incorporates GPT 3.5 as a sentiment analysis component, to reduce user dissatisfaction in this service process. To validate the proposal, experiments were conducted with 50 internet and cable television service owners to evaluate satisfaction and accuracy in recognizing their emotions. The results indicated that 86% of the respondents were satisfied with the chatbot service, and the satisfaction index reached 77.9, surpassing the minimum threshold of 75 points for providing quality customer service established by the industry. The methodology behind these results is detailed in the following research.
Download

Paper Nr:	33
Title:	Roadmap for Implementing Business Intelligence Systems in Higher Education Institutions: Validation of a Case Study at the University of Trás-os-Montes and Alto Douro
Authors:	Romeu Sequeira, Arsénio Reis, Frederico Branco and Paulo Alves
Abstract:	The adoption of effective policies and access to relevant information are critical to improving strategic management and performance monitoring in Higher Education Institutions (HEIs), which is essential to promote data-driven decision-making. This article describes how an HEI can carry out the validation process when implementing a Business Intelligence (BI) system, and provides a detailed guide to doing so. Through a case study at the University of Trás-os-Montes and Alto Douro, a structured roadmap is validated, which acts as a visual and sequential guide to facilitate the effective implementation of BI solutions. The validation carried out through semi-structured interviews with experts and evaluation of dashboards by user groups, not only confirms the applicability and efficiency of the proposed model but also emphasises its practical relevance, providing valuable insights for the adaptation and use of BI in different educational contexts.
Download

Short Papers

Paper Nr:	18
Title:	Artificial Neural Network Model for Predicting Excavator Downtime
Authors:	Vesna S. Brkić, Ivan Mihajlović, Martina Perišić and Nemanja Janev
Abstract:	Previous research shows the significance of maintenance in enhancing performance levels and reducing system costs of equipment. This paper aims to develop a quantitative model for predicting the failure rate of excavators using artificial neural networks (ANN). As an input to the ANN, the duration times of 590 excavator downtimes measured over 198 days at the mining site in Serbia were used to obtain a classification of failures longer than an hour based on the previous 14 days, in aim to prevent potential indirect financial losses which could be over 15000€/hour. A Pareto analysis of the observed data was also performed and showed the technological type of downtime as the most frequent. The results show that the ANN modeling is suitable for mapping the non-linear relationship between excavation activities and the failure rates of excavators. The results showed that the proposed ANN model provides an accurate estimating tool for the early planning stage to predict failure rates of excavators. Future research avenue proposal is directed at monitoring and forecasting the exact duration of excavator downtime in real-time.
Download

Paper Nr:	26
Title:	Business-RAG: Information Extraction for Business Insights
Authors:	Muhammad Arslan and Christophe Cruz
Abstract:	Enterprises depend on diverse data like invoices, news articles, legal documents, and financial records to operate. Efficient Information Extraction (IE) is essential for extracting valuable insights from this data for decision-making. Natural Language Processing (NLP) has transformed IE, enabling rapid and accurate analysis of vast datasets. Tasks such as Named Entity Recognition (NER), Relation Extraction (RE), Event Extraction (EE), Term Extraction (TE), and Topic Modeling (TM) are vital across sectors. Yet, implementing these methods individually can be resource-intensive, especially for smaller organizations lacking in Research and Development (R&D) capabilities. Large Language Models (LLMs), powered by Generative Artificial Intelligence (GenAI), offer a cost-effective solution, seamlessly handling multiple IE tasks. Despite their capabilities, LLMs may struggle with domain-specific queries, leading to inaccuracies. To overcome this challenge, Retrieval-Augmented Generation (RAG) complements LLMs by enhancing IE with external data retrieval, ensuring accuracy and relevance. While the adoption of RAG with LLMs is increasing, comprehensive business applications utilizing this integration remain limited. This paper addresses this gap by introducing a novel application named Business-RAG, showcasing its potential and encouraging further research in this domain.
Download

Paper Nr:	38
Title:	Gathering and Matching Data from the Web: The Bibliographic Data Collection Case Study
Authors:	Olga Cherednichenko, Lubomir Nebesky and Marián Kováč
Abstract:	As a result of the analysis of existing approaches to consolidating data on research activities we highlight a number of issues. Firstly, the automating the process of data collection which is included the comparing data from different sources. Secondly, the use of external services to obtain bibliographic information which is accompanied by the receipt of erroneous data. The idea of a tracking system for research activity implies that we collect and consolidate data from different web sources and keep them in order to provide relevant bibliographic information. We outline several key points to consider different spellings of the authors’ names, data duplication, and filtering out erroneously data. The purpose of the study is to improve the accuracy of comparing bibliographic data from different indexing systems. We propose the framework for gathering and matching bibliographic data from the web. The experimental results show the performance of the proposed algorithm with reaching 0.88 for the F1 metric. The software prototype is developed. The ways to improve the proposed algorithm have been identified, which opens up opportunities for further research.
Download

Paper Nr:	32
Title:	Exploring the Test Driven Development of a Big Data Infrastructure Examining Gun Violence Incidents in the United States of America
Authors:	Daniel Staegemann, Malte Rathjens, Hannes Hinniger, Vivian Schmidt and Klaus Turowski
Abstract:	Big data (BD) and the systems used for its harnessing heavily impact many aspects of today’s society and it has been repeatedly shown that they can positively impact the operations of organizations that incorporate them. However, creating and maintaining these applications is extremely challenging. Therefore, it is necessary to pay additional attention to the corresponding quality assurance. One software engineering approach that combines high test coverage, the enabling of comprehensive regression tests, but also positively impacts the developed applications’ design, is test driven development. Even though by now it has a somewhat long history in software development in general, its use in the context of BD engineering is not common. However, an approach for the test driven development of BD applications that is based on microservices has been proposed rather recently. To gain further insights into its feasibility, the publication at hand explores its application in the context of a prototypical project implementation. Hereby, the chosen use case is the analysis and prediction of gun violence incidents in the United States of America, which also incorporates NFL match game data, under the assumption that the games could potentially influence the occurrence of such incidents.
Download

Area 2 - Business Models and Business Processes

Short Papers

Paper Nr:	8
Title:	Hidden Champions Revised: Towards a New Conceptual Framework
Authors:	Daniela Podevin, Laura Bies, Tobias Greff and Dirk Werth
Abstract:	The business environment for Hidden Champions has evolved dramatically, with rapid technological advancements and the rise of Artificial Intelligence (AI) accelerating change and increasing the importance of data. This paper explores the need for an updated Hidden Champions model that reflects the critical roles of innovation, technology utilization, and social media visibility, emphasizing the transformative impact of AI on competitiveness. It suggests that mastery of AI and effective data use are indispensable for companies seeking to secure and expand their market positions in the current digital era. Integrating these elements, alongside a robust social media presence, is proposed as an essential criterion for identifying Hidden Champions. The paper argues that these updated core elements can serve as helpful indicators of a company’s potential for growth and competitive advantage, highlighting that firms at the forefront of AI and data analytics can outperform rivals, regardless of their size or global footprint. It concludes that adapting to these changes by embracing AI, leveraging data, and engaging with social media can significantly enhance a company’s ability to innovate and remain competitive in the fast-evolving market landscape. The Hidden Challenger model presented in this study proposes a strong theoretical basis, explaining the core elements and enabling future research to empirically validate the innovative model. It aims to guide companies and policymakers toward strategies that bolster competitiveness and ensure sustainable growth in an age dominated by digital transformation and AI.
Download

Paper Nr:	36
Title:	Process Mining Enabled Cognitive RPA to Automate Data Entry Tasks in ERP Systems
Authors:	Ali Suleiman and Gamal Kassem
Abstract:	Manual data entry tasks are a common source of inefficiency in organizations. They are time-consuming and prone to errors. RPA is an emerging automation approach that is non-intrusive and mimics the user interaction with interfaces. It could be used along with an AI tool such as NLP to automate data entry tasks that come from unstructured sources such as emails. This paper proposes a cognitive RPA framework that utilizes process mining to find, select and configure cognitive RPA to automate data entry tasks to ERP systems. Process mining is used to extract process models generated from event logs to allow for the accurate analysis of tasks. The framework also utilizes the concept of UI logs for the configuration steps. A proof of concept NLP model was developed, and the framework was evaluated by an expert to validate its potential and highlight areas for improvement. While limitations exist, such as the small NLP training dataset, the paper demonstrates the potential of this approach for improving efficiency, reducing errors, and enhancing decision-making in organizations. Future work includes expanding the training data and exploring user interface design for error correction.
Download

Paper Nr:	37
Title:	A Systematic Review to Identify Patterns Types and Analysis Objectives for the Discovery of Business Rules from Event Logs Using Machine Learning
Authors:	Menna Wael and Gamal Kassem
Abstract:	Business processes are structured and executed based on business rules. Information systems executing the business processes store the execution data in event logs. The event logs can be analyzed using machine learning algorithms to discover business rules in the business process execution. In which various algorithms can be applied to event log data to discover rules/patterns related to the business process. The application of machine learning on event log data to discover the business rules needs extensive process mining expertise and knowledge from the process analyst; therefore, there is a need to facilitate the application of machine learning on event log data to reach different analysis objectives. This can be done through identifying the pattern types related to performing different machine learning tasks on event log data, and the different analysis objectives for the discovery of business rules from event logs. However, it was found that no systematic review was previously conducted to collect this information; therefore, the focus of this paper is to conduct a systematic review to collect from research the different pattern/rule types within the event log data that can be discovered and the different analysis objectives for the discovery of business rules from event logs.
Download

Area 3 - Technologies and Applications

Full Papers

Paper Nr:	30
Title:	Classification of Peruvian Elementary School Students with Low Achievement Problems Using Clustering Algorithms and ERCE Evaluation
Authors:	Nancy Rojas-Salvatierra, Lucas Parodi-Roman and Peter Montalvo
Abstract:	At present there are several problems that affect students and their academic performance such as low socioeconomic status that can cause lack of resources both in their homes and in the school. In addition to psychological and personal problems in which students can be involved. According to various national and international examinations the academic level in Peru is quite low because the problems mentioned above are difficult to identify, it is not possible to propose a viable solution, which is why we propose a Machine Learning model based on Clustering algorithms such as KMeans, Birch and Aglomerative that manage to group students by the most relevant characteristics or disadvantages they present.
Download

Short Papers

Paper Nr:	10
Title:	Enhancing Returns Management in Fashion E-Commerce: Industry Insights on AI-Based Prediction and Recommendation Systems
Authors:	Soeren Gry, Marie Niederlaender and Dirk Werth
Abstract:	The fashion industry is one of the most problematic sectors in terms of sustainability. The fashion e-commerce sector is experiencing a surge in sales, which is leading to a significant increase in returns. This, in turn, is placing a considerable burden on the environment. High transport volumes or even the destruction of garments through returns pose major environmental and also economic problems. This study is based on a survey and expert interviews with decision-makers from the fashion industry. It provides indications of how an AI-based prediction and recommendation system could be used to avoid returns and manage them in an ecologically and economically sensible way. On the one hand, use cases are discussed that can be applied in the webshop system before the customer places an order, and on the other hand, ways are shown how returns predictions can support planning in the reverse logistics network.
Download

Paper Nr:	19
Title:	Framework for Modeling the Propagation of Disturbances in Smart Construction Sites
Authors:	Ali Attajer and Boubakeur Mecheri
Abstract:	The construction sector is currently undergoing a paradigm shift by technological advances. This transformation has led to the emergence of the concept of “Construction 4.0”. However, despite these advances, improving resilience - the ability to adapt effectively to unexpected events - remains a major challenge. In this work, we aim to bridge this scientific gap by proposing a framework to systematically characterize and model disturbances and their propagation. We instantiate the framework in a case study using discrete event simulation in FlexSim. In this model, we simulate a smart construction site where construction activities are automated by intelligent and autonomous entities, such as robots, automated guided vehicles, and autonomous cranes. Moreover, we examine two scenarios to understand how a type of disturbance, with specific characteristics, propagates through the system and impacts the continuity of construction activities and operations. The results provide essential insights into the impact of disturbances on work progress, project duration, the capacities of autonomous entities, and stock levels.
Download

Paper Nr:	27
Title:	Speech Recognition for Inventory Management in Small Businesses
Authors:	Bruno Tiglla-Arrascue, Junior Huerta-Pahuacho and Luis Canaval
Abstract:	In recent years, we have seen an increase in independent businesses working primarily focused on online sales, where they offer products through ads and manage the business with electronic tools. This could leave behind some traditional businesses, especially those that are managed by a single family, where the adaption of new technologies is slower than new business. That’s why we want to give them a tool that it’s easy to control, a virtual assistant where they can manage the inventory even if they don’t know about databases. For this work, we propose to create a speech-to-text platform with machine learning so those users who have difficulties adapting to these new tools can use their voice to command the database and have first contact with these new technologies. Through a fine-tuning process to a pre-trained speech-to-text model in Spanish, we managed to obtain a percentage error result lower than the model used, this being 14.3%, this means that our model has a better accuracy in the context of a Peruvian convenience store.
Download

Paper Nr:	34
Title:	A Bibliometric Analysis of Green Accounting, Environmental Accounting and Green Business Publications in a Global Perspective
Authors:	Hamide Özyürek
Abstract:	This study aims to conduct a comprehensive review of research on green accounting, environmental accounting and green business. The methodology employs advanced bibliometric techniques such as co-citation analysis, trend topics, thematic evolution. A total of 1603 documents from the Web of Science, spanning the period between 1991 and April 23, 2024, were screened and analyzed using R program. The findings revealed six thematic clusters: Social and environmental accounting, emergy, green business, green innovation, environmental accounting, and green accounting. The most cited authors are Boyd and Banzhaf, Cho and Patten, and Laufer. The findings indicates that the journals with the highest number of articles and citations in this field are Journal of Cleaner Production, Sustainability, Ecological Economics, and Accounting, Auditing & Accountability Journal. When considering the number of articles and citations by country, China, the USA, and Italy emerge as the leading contributors.
Download