KMIS 2024 Abstracts

Full Papers Short Papers
Full Papers
Paper Nr: 44
Title:

Scientific Claim Verification with Fine-Tuned NLI Models
Authors:

Miloš Košprdić, Adela Ljajić, Darija Medvecki, Bojana Bašaragin and Nikola Milošević
Abstract: This paper introduces the foundation for the third component of a pioneering open-source scientific question-answering system. The system is designed to provide referenced, automatically vetted, and verifiable answers in the scientific domain where hallucinations and misinformation are intolerable. This Verification Engine is based on models fine-tuned for the Natural Language Inference task using an additionally processed SciFact dataset. Our experiments, involving eight fine-tuned models based on RoBERTa Large, XLM RoBERTa Large, DeBERTa, and DeBERTa SQuAD, show promising results. Notably, the DeBERTa model fine-tuned on our dataset achieved the highest F1 score of 88%. Furthermore, evaluating our best model on the HealthVer dataset resulted in an F1 score of 48%, outperforming other models by more than 12%. Additionally, our model demonstrated superior performance with a 7% absolute increase in F1 score compared to the best-performing GPT-4 model on the same test set in a zero-shot regime. These findings suggest that our system can significantly enhance scientists’ productivity while fostering trust in the use of generative language models in scientific environments.
Download
Paper Nr: 56
Title:

Semantic Support Points for on the Fly Knowledge Encoding in Heterogenous Systems
Authors:

Daniel Spieldenner, André Antakli, Torsten Spieldenner and Harkiran Sahota
Abstract: Connecting participants in heterogenous environments and allowing for seamless interaction is a challenging task that requires profound knowledge of the system envolved and data exchanged. The knowledge on how to integrate and use certain systems is usually given in the form of manuals, code documentation or needs to be derived by the developer by understanding and interpreting source code or data sets. With this work, we propose an approach to provide this knowledge about data via so called semantic support points, making use of Semantic Web technologies and appropriate ontologies, in a resource efficient fashion by creating semantic representations only when and where needed. Instead of just providing semantic meta data to describe actors in the environment, we make it possible to embed actual data values from data sources like databases and payloads exchanged between systems into a virtual semantic cloud, allowing for interactions purely on the semantic data representations and propagating computation results back to the system layer automatically.
Download
Paper Nr: 66
Title:

Values and Enablers of Lessons Learned Practices: Investigating Construction Industry Context
Authors:

Jeffrey Boon Hui Yap
Abstract: In the realm of construction project management, the value of "lessons learned (LL)" cannot be overstated. LL, as an important approach for effective project management and continuous improvement, is analysed in this study, with the aim to advance the impact of LL by determining the values of LL practices and examining the enablers that positively influence LL practices in the construction industry. A detailed literature review has revealed nine (9) values and seven (7) enablers of LL practices relevant to the construction industry’s context. Using a questionnaire survey involving 129 Malaysian construction professionals selected based on non-probability techniques, the significance of the values and enablers is prioritised based on mean scores. Findings reveal that LL practices help to avoid making similar past mistakes, optimize project performance and engender collaborative learning in the project team. Individual-related enablers are perceived to be more influential than organisational-related enablers in implementing LL in construction projects. Collective and conscious efforts in fostering a learning culture are crucial to encourage the construction industry to embrace LL practices and help individuals and organisations thrive.
Download
Paper Nr: 75
Title:

Evaluation of the Contribution of Knowledge Management to Efficiency in the Manufacturing Industry Through Machine Learning
Authors:

Juan Ibujés-Villacís
Abstract: Knowledge management (KM) has been instrumental for organizations to improve their efficiency. The objective of this research is to determine the contribution of knowledge management (KM) to manufacturing industry efficiency, using machine learning models to predict the relevant KM factors that should be taken into account to improve efficiency. Given the quantitative nature of the research, in the first phase, data on variables associated with KM factors and efficiency were collected and processed. In the second phase, four supervised machine learning models were developed to predict which manufacturing companies are efficient in their production process based on a set of KM factors. The study was based on information from 142 manufacturing companies in the province of Pichincha, Ecuador. The results show that the relevant KM factors that contribute to business efficiency are policies and strategies, organizational structure, technology, incentive systems and organizational culture. This pioneering study in Ecuador allows predicting the relevant KM factors that impact the efficiency of manufacturing firms. This article contributes to the field of knowledge management and provides information on the KM factors that manufacturing firms should focus on to achieve greater efficiency.
Download
Paper Nr: 90
Title:

Charting the Transformation of Enterprise Information Management: AI Explainability and Transparency in EIM Practice
Authors:

Lufan Zhang and Paul Scifleet
Abstract: Today’s data-intensive environment poses significant challenges for enterprises in managing their vital information assets that often exceed manual capabilities. Despite a promising potential to assist, there’s mistrust and misunderstanding of the values AI presents to Enterprise Information Management. This paper investigates the current state of AI-led changes to EIM practices and proposes an approach to improve understanding of AI’s transformative role and impact on EIM. By charting AI use in EIM platforms across five areas - AI development, AI techniques, AI-integrated EIM capabilities, AI applications, and AI impacts – along with practice-based criteria for evaluating AI-integrated EIM solutions, this paper lays the foundation for explainable and transparent AI in EIM.
Download
Paper Nr: 107
Title:

Uncertainty Analysis in Population-Based Dynamic Microsimulation Models: A Review of Literature
Authors:

Miia Rissanen and Jyrki Savolainen
Abstract: This paper reviews population-based dynamic microsimulation (DMs) models used in policy analysis and decision support of social systems and demographics. The application of uncertainty analysis (UA) methods is examined focusing on how probabilistic Monte Carlo (MC) simulation technique is being used and reported. Secondly, inspired by the expanding possibilities of data, this analysis examines the models' capability to uncover finer temporal variations beyond traditional yearly intervals and the use of near real-time data in the reported studies. The analysis of the 44 studies included in this preliminary literature review reveals a lack in the rigorous application of UA and transparent communication of results, particularly in the social sciences. Despite the advances of data availability and modeling, no research attempts were found that would indicate a shift of paradigm from historical data-driven models to real-time data. It is suggested that DM studies in this context could benefit from some mutually agreed standardized reporting guidelines for UA. This literature review serves as a preliminary exploration of the topic, highlighting the need for a more comprehensive and systematic survey to thoroughly assess the current state of research.
Download
Paper Nr: 129
Title:

Knowledge Pyramid Perspective of the Political Data Ecosystem: A Case Study of Bhutan
Authors:

Phub Namgay and Pema Wangdi
Abstract: This study examines the dynamics of data management and knowledge flow in the political data ecosystem through the lens of the Knowledge Pyramid. We used open-government electoral documents and polling data for granular insights into how data, information, knowledge, and wisdom (DIKW) are managed in Bhutan’s political data ecosystem. Bhutan’s electoral stakeholders and political parties manage and use DIKW of varying types, sizes, and complexities. In particular, political parties use information systems, websites, and social media to manage data and construct and use knowledge for political activities. Democracy is still young and gaining a foothold in Bhutan. The political parties do not employ complex data technologies and rich human resources to manage DIKW emanating from the political data ecosystem. Thus, scope exists for electoral stakeholders and political actors to explore and adopt effective and efficient knowledge management infras-tructures to deal with DIKW elements in the political arena, namely the complex dynamics of turning raw data into higher elements of the Knowledge Pyramid. In addition to contributing to the knowledge management literature through an in-depth account of the DIKW aspects in the political space, this paper demonstrated the analytical and explanatory power of the Knowledge Pyramid for discourse on the political data ecosystem.
Download
Paper Nr: 135
Title:

Support Learning Design and Analytics with EduP Knowledge Model
Authors:

Thi P. Nguyen, Ngoc T. Nguyen-Huynh and Thi H. Vu
Abstract: Learning design (LD) have been a prominent topic in the academic community for many years. It aims at planning and organizing learning activities and resources to promote learning process and engage students in achieving learning outcomes. Learning analytics (LA) has matured in the education field and developed a strong connection with learning design. Learning analytics provides valuable insights to inform learning design decisions, while learning design serves as a means to turn learning analytics results into actionable strategies. Their alignment completes the big picture for enhancing teaching and learning. Despite numerous studies proposing means to support LD/LA and their alignment, both fields still face many challenges due to the lack of a consolidated framework for reflecting on the various types of knowledge essential for LD/LA. This paper aims at proposing a comprehensive framework, named EduP (Education-Domain-User-Pedagogy), that supports LD/LA by leveraging different types of knowledge. The main contributions of the framework include a knowledge model and an insight engine. The knowledge model helps clarify essential components for LD/LA and their relationships, while the insight engine addresses how this knowledge is accessible to teachers in the context of LD/LA. A brief discussion on the implications and future research is also presented.
Download
Paper Nr: 140
Title:

Crafting the Future: Developing and Evaluating a Digital Mindset Competence Model for the Industrial Craft Sector
Authors:

Seyma Kocak and Jan Pawlowski
Abstract: The recent development of digitization has significantly influenced various sectors of the economy, and the Industrial Craft Sector is no exception. The transition to digital technologies and processes is inevitable and holds the potential for increasing efficiency and creating competitive advantages. This research used the Design Science methodology to develop a Digital Mindset Competence Model. This model comprises eight dimensions specifically tailored to the requirements and challenges of the Industrial Craft Sector. These dimensions aim to promote and strengthen the digital mindset among professionals in the Industrial Craft Sector. To ensure the validity and relevance of this model, experts from the Industrial Craft Sector were involved in a qualitative methodology. The combination of scientific methodology and practical experience ensures a comprehensive perspective and guarantees the applicability of the developed model. The results of this research underscore the importance of digital transformation in the Industrial Craft Sector and the necessity of a digital mindset. The developed Digital Mindset Competence Model provides a targeted approach to promoting digital competencies in the Industrial Craft Sector and guides future developments in this area. It becomes evident that an appropriate digital mindset is essential to optimally leverage the potentials of digitization in the Industrial Craft Sector and successfully navigate continuous change. This scholarly contribution contributes to raising awareness of the significance of a digital mindset in the Industrial Craft Sector. It forms a basis for further investigations and practical applications within digital transformation.
Download
Paper Nr: 149
Title:

Using Formal Concept Analysis for Corpus Visualisation and Relevance Analysis
Authors:

Fabrice Boissier, Irina Rychkova and Benedicte Le Grand
Abstract: Corpora analysis is a common task in digital humanities that profits from the advances in topic modeling and visualization from the computer science and information system fields. Topic modeling is often done using methods from the Latent Dirichlet Allocation (LDA) family, and visualizations usually propose views based on the input documents and topics found. In this paper, we first explore the use of Formal Concept Analysis (FCA) as a replacement for LDA in order to visualize the most important keywords and then the relevance of multiple documents concerning close topics. FCA offers another method for analyzing texts that is not based on probabilities but on the analysis of a lattice and its formal concepts. The main processing pipeline is as follows: first, documents are cleaned using TreeTagger and BabelFy; next, a lattice is built. Following this, the mutual impact is calculated as part of the FCA process. Finally, a force-based graph is generated. The output map is composed of a graph displaying keywords as rings of importance, and documents positioned based on their relevance. Three experiments are presented to evaluate the keywords displayed and how well relevance is evolving on the output map.
Download
Paper Nr: 157
Title:

Building Atlas of Knowledge Maps: Towards Smarter Collaboration
Authors:

Anna Kuznetsova, Tatiana Gavrilova and Olga Alkanova
Abstract: The paper discusses the possibilities and prospects for creating corporate atlas of knowledge maps – a visual guide of diagrams describing the intellectual assets of the enterprise. The discussed case is based on the university business school. Mapping or visualization provides information transparency of communications in universities making collaboration smart and effective. The walls of universities are opaque, and visualization provide a higher level of teaching, research, consulting and administration. The paper presents the preliminary results of the project “Methodology and technology for developing digital knowledge maps for education and research teams’’ and proposes and describes specific features of a systematic repository of diagrams, that is called an atlas of knowledge maps. We developed a set of diagrams to describe knowledge, created an ontology of the properties of such maps and suggested considering the most popular ones as a kind of atlas from which decision makers can select relevant maps for their work. The survey is preceded by the use of ontologies - conceptual models of areas of knowledge and professional activities of the teacher. In general, the approach can be adapted to business companies and government organizations if they are interested in disclosing their intellectual capital.
Paper Nr: 166
Title:

Benchmarking of Retrieval Augmented Generation: A Comprehensive Systematic Literature Review on Evaluation Dimensions, Evaluation Metrics and Datasets
Authors:

Simon Knollmeyer, Oğuz Caymazer, Leonid Koval, Muhammad U. Akmal, Saara Asif, Selvine G. Mathias and Daniel Großmann
Abstract: Despite the rapid advancements in the field of Large Language Models (LLM), traditional benchmarks have proven to be inadequate for assessing the performance of Retrieval Augmented Generation (RAG) systems. Therefore, this paper presents a comprehensive systematic literature review of evaluation dimensions, metrics, and datasets for RAG systems. This review identifies key evaluation dimensions such as context relevance, faithfulness, answer relevance, correctness, and citation quality. For each evaluation dimension, several metrics and evaluators are proposed on how to assess them. This paper synthesizes the findings from 12 relevant papers and presents a concept matrix that categorizes each evaluation approach. The results provide a foundation for the development of robust evaluation frameworks and suitable datasets that are essential for the effective implementation and deployment of RAG systems in real-world applications.
Download
Short Papers
Paper Nr: 23
Title:

A Model for Designing Personalized and Context-Aware Nudges
Authors:

Randi Karlsen and Anders Andersen
Abstract: Nudging is a popular approach used to influence people to change their behavior towards a desirable goal. To be effective, nudges should be tailored to the user’s specific needs based on user profile, current behavior, and context information. In this paper, we target the questions of what to nudge for and when to nudge, and how nudges can be automatically designed at a time when the user needs nudging. We present a model for personalized and context-aware nudge design that is adaptive in that it continuously tailors the set of relevant activities and the time for creating a nudge to the user’s needs for behavioral change. The model follows a just-in-time approach, where nudges are created at the time when nudging is needed, based on the user’s current situation.
Paper Nr: 33
Title:

Automatic Transcription Systems: A Game Changer for Court Hearings
Authors:

Alan Lyra, Carlos E. Barbosa, Herbert Salazar, Matheus Argôlo, Yuri Lima, Rebeca Motta and Jano Moreira de Souza
Abstract: Lawsuits typically require a long time for resolution, and many court hearings may occur during the trial process. Legally, both parties must transcribe them and be open to the public if desired by the court. In court proceedings, a transcript is a record of all judges’ decisions, the spoken arguments by the lawyers, and the depositions of the defense and witnesses. The scenario in Brazil is that for a long time, this process was manual, with a person responsible for the typing transcription. Today, with the electronic process, the court does not provide typed transcriptions anymore, but instead, the audio or video recordings of the hearings. In our work, we developed an automatic transcription solution for court hearings to obtain the best possible transcription considering current technologies’ limitations, recording quality, participants’ diction, and commonly used jargon in the legal sphere. With this work, we expect to ease this burdensome task with technical support and have a direct contribution to the legal environment.
Download
Paper Nr: 46
Title:

Data Governance to Be a Data-Driven Organization
Authors:

Carlos A. Bassi, Solange N. Alves-Souza and Luiz Sergio de Souza
Abstract: Many organizations have been trying to become data-driven which their business decisions, their relationships with customers / suppliers, the innovation of their products / services, the improvement in their performance and their growth are based on the collection and analysis of an increasing volume of data. To reach this level, organizations need to overcome a series of challenges related to the way they govern data and creating a data-driven culture. The main challenges to be overcome are directly related to data culture and the culture of the organization itself. These paper presents the results of a survey performed among 67 professionals with experience in Data Governance (DG) in which was possible to identify the main challenges to establish a DG program and data-driven culture in organizations, besides priories actions to face these challenges. The challenges and necessary actions to implement a DG program are shown and discussed. Addressing these challenges is fundamental to raising the organization culture and maturity in DG and, consequently, becoming a Data-driven organization.
Download
Paper Nr: 74
Title:

Design and Evaluation of Microteaching: Emergent Learning for Acquiring Classroom Management Skill in Teacher Education
Authors:

Dai Sakuma, Keitaro Tokutake and Masao Murota
Abstract: Schools in countries struggling with academic achievement gaps need to improve the teaching and support skills for students who facilitate classes in these gaps. This study focused on methods for acquiring complex classroom management skills for pre-service teachers. The aim of the study was to design and validate a method for teacher candidates to learn these behaviors. To achieve this, microteaching sessions in which unexpected behaviors occur were designed and carried out. A video recording of the microteaching was used in the evaluation experiment. Evaluators were five expert teachers in Japan. Statistical tests using the results of the questionnaire responses revealed that the simulated situations by student roles were close to actual situations with real students. It was also confirmed that the teaching candidates experienced situations that required various management behaviors. These results indicate that the microteaching sessions designed by the authors are useful as a method for emergent learning to achieve management skills in the classroom to control unexpected behaviors.
Download
Paper Nr: 83
Title:

Exploring Centralized, Decentralized, and Hybrid Approaches to Micro-Credential Issuance in HEI Alliances
Authors:

Padmasheela Kiiskilä and Henri Pirkkalainen
Abstract: Micro-Credentials (MCs) are seen as a way by Higher Education Institutions (HEIs) to equip learners with the essential skills for their careers or professional development. In Europe, HEIs are joining forces to form alliances to offer a broad range of MCs and make them tamper-proof, verifiable, and shareable. Although extensive research is being done on MCs, there is a major research gap in identifying and comparing different ways alliances can manage issuance of MCs. We identified two approaches in practice and through a case study, identified a third approach alliances can utilize. This paper also addresses this gap by using the data governance contingency model to provide a comparison of all three approaches that alliances can utilize in selecting the most suitable one for their business strategy. To achieve this, a qualitative case study is conducted with in-depth interviews with administrators from HEIs that are partners of an alliance. This study contributes to the governance of MCs through identification and comparison of the three approaches - centralized, decentralized, and hybrid in the context of MC issuance by HEI alliances.
Download
Paper Nr: 84
Title:

Evaluating Healthcare Automation: A Multi-Case Study on the Utilization of Automation Initiatives in Healthcare Operations
Authors:

Jani Kaitosalmi and Milla Ratia
Abstract: Automation technologies such as robotic process automation (RPA) and intelligent automation (IA) are essential for managing rising healthcare costs and ensuring sustainable health services. Although these solutions have been implemented in several Finnish healthcare organizations, their overall impact has not been systematically evaluated. This research investigates the impact and evaluation of healthcare automation through a multi-case study conducted in two Finnish healthcare organizations. While automation has improved resource utilization, process efficiency, and standardization across units, the findings highlight the need for a more comprehensive evaluation and continuous monitoring of automation benefits. Future research should focus on developing a specific evaluation framework tailored to healthcare automation technologies. The adoption of holistic evaluation methods could allow healthcare organizations to better understand the impact of automation and further enhance operational efficiency and patient care.
Download
Paper Nr: 95
Title:

Digital Transformation of B2B Sales Processes
Authors:

Laura Lahtinen, Tommi Mahlamäki and Jussi Myllärniemi
Abstract: This paper explores the impact of digital transformation on sales practices. It emphasizes how businesses adopt new digital technologies to enhance efficiency, customer engagement, and overall business outcomes. Digital transformation involves both digitalizing analog processes and leveraging innovative technologies like artificial intelligence and automation in sales operations. Some companies have successfully integrated digital tools to improve B2B sales effectiveness and profitability. However, challenges persist, including organizational resistance and the need for updated managerial practices and performance metrics. The main contribution of this paper is the identification of changes in sales caused by digital transformation. Despite these changes and challenges, digital transformation in sales signifies a shift towards more consultative and adaptive sales approaches. This shift necessitates a re-evaluation of traditional sales practices and a proactive embrace of technological advancements for sustained competitive advantage in dynamic markets.
Download
Paper Nr: 96
Title:

Sales Development in Business-to-Business Markets
Authors:

Tommi Mahlamäki and Juho Martikainen
Abstract: Sales development has emerged as a critical strategy within the business-to-business (B2B) landscape, propelled by the ongoing digitalization of sales processes. This paper explores how digital technologies, including artificial intelligence (AI), machine learning, CRM systems, and e-commerce platforms, have reshaped sales strategies and interactions between buyers and sellers. As digital tools become integral to modern sales operations, the role of sales development in generating leads and fostering customer relationships is increasingly prominent. The paper reviews existing literature on sales development, highlighting its evolution from traditional sales models to contemporary digital frameworks. Furthermore, the study proposes an initial conceptual framework for understanding sales development in B2B contexts, setting the stage for empirical research to explore its implementation and effectiveness in enhancing sales performance and customer engagement.
Download
Paper Nr: 106
Title:

Importance of Context Awareness in NLP
Authors:

Nour Matta, Nada Matta and Philippe Herr
Abstract: Context is a complex notion, that enables the understanding of happenings and concepts in an environment and the analysis of their influence (Adomavicius et al, 2011) As previously mentioned, context plays a major role in assigning meanings to words, sentences, and texts when dealing with text analysis. Multiple natural language processing approaches aim to consider “context” in analyzing the information extracted and applying a sort of word sense disambiguation (Adhikari et al, 2019). Numerous intelligence systems require knowledge of happening and are context dependent, but the definition of context and context elements used varies from one application to another based on needs. Context plays several roles in text analysis especially to reduce ambiguity and semantic extraction. In this paper, main influence of context on TextMining and NLP are shown.
Download
Paper Nr: 112
Title:

How Organizational Improvisational, Transformational Leadership Styles Impact Innovation Performance of Start-Up Companies in VUCA Environments
Authors:

Wenxi Xiong and Yongzhong Yang
Abstract: Entering the competitive VUCA environment, the traditional management model with prediction and control as the main measures has not adapted to the needs of the times in certain situations, and it has become a hot topic for start-ups to survive and grow in the current unpredictable environment and achieve breakthrough innovation. This study constructs a correlation model between organizational improvisation, transformational leadership, and innovation performance of start-ups, and uses the VUCA environment as a moderating variable. The findings confirm that all dimensions of organisational improvisation and transformational leadership significantly affect the innovation performance of start-ups; Organizational improvisation and transformational leadership positively interact to influence innovation performance of start-ups, i.e. organizational improvisation and transformational leadership reinforce each other's influence on innovation performance of start-ups; The VUCA environment positively moderates the impact of organizational improvisation and transformational leadership on innovation performance of start-ups. This study will help start-ups to fully grasp the fleeting opportunities and respond to the changing external environment in a timely manner to further enhance the competitive advantage of start-ups, and provide practical guidance for top managers of start-ups to enhance their innovation performance.
Download
Paper Nr: 115
Title:

Knowledge Sharing in Financial Institutions to Assist with IT Service Management: A Thematic Analysis
Authors:

Cornelius J. Niemand and Josef Langerman
Abstract: The applications and services provided by financial institutions are important to individuals and economies. These applications and services are fragile because of service failures that are inherent in technology. The purpose of this article is to show how knowledge management can mitigate service disruption in financial institutions. By using bibliometric analysis and a structured literature review based on the PRISMA 2020 guidelines, we identified five major themes that drive knowledge management (KM) practices in information technology (IT) management in financial institutions. These themes identified are centered on the organizational environment, the motivation of employees, the people profile for example gender and race, and lastly the use of technology. By bolstering these KM practices in the IT service management (ITSM) of financial institutions we hope to shorten the time between system failures and shorten the actual time to repair failures. Knowledge management in IT management and especially ITSM is under-researched in financial institutions, and the KM themes identified provide some signposts to improved collaboration and better theorisation.
Download
Paper Nr: 125
Title:

Performing Entity Relationship Model Extraction from Data and Schema Information as a Basis for Data Integration
Authors:

Philipp Schmurr, Andreas Schmidt, Karl-Uwe Stucky, Wolfgang Suess and Veit Hagenmeyer
Abstract: The goal of this work is to allow domain experts to properly perform data integration themselves and not to rely on external resources. This way the long-term data integration quality is not endangered and therefore cost for external resources can be saved. To achieve this, we propose a new approach that enables data integration based on entity-relationship (ER) models derived from arbitrary data sources. ER models are abstract and simply define all entities and relations needed for integration, which makes them easy to understand. Strategies to extract ER models from various standard data sources - relational databases, XML files and OWL data are presented and a concept on how to extend it to arbitrary other data sources is introduced. Furthermore, the extracted models are a foundation to perform graphical data integration into an ontology based model and, thus, contribute to a harmonized knowledge management in heterogeneous data and information environments. It can be summarized as a strategy to improve the interoperability of existing data according to the FAIR principles.
Download
Paper Nr: 127
Title:

FAIRlead: A Conceptual Framework for a Model Driven Software Development Approach in the Field of FAIR Data Management
Authors:

Andreas Schmidt, Mohamed A. Koubaa, Nan Liu, Philipp Schmurr, Karl-Uwe Stucky and Wolfgang Süß
Abstract: The publication of scientific results together with the underlying experiments is an important source of further research. In 2016, the “FAIR Guiding Principles for scientific data management and stewardship” were published, in which the authors postulate a series of guidelines for improving the (F)indability, (A)ccessibility, (I)nteroperability and (R)eusability of digital information (FAIR). The point (I)nteroperability deals with the prerequisites for the reusability of digital objects. The central point here is the need to have a common understanding of the meaning of digital objects. This understanding is provided by formal languages of knowledge representation (ontologies), which describe the actual data. These descriptions of data are also known as metadata. As part of our current work at the Institute for Automation and Applied Computer Science (IAI) at KIT, we are implementing novel concepts and technologies for the sustainable handling of research data using high-quality metadata. As part of this work, we plan to develop a software tool that can be used to enrich data with suitable metadata and thus automate the process of making research results available. A key requirement is that the tool must be independent of the underlying domain. In order to be able to deal with data from any domain, we have opted for a model-driven approach in which an ontology, and possibly other platform-specific information, are input for a software generator, which then generates an (interactive) tool for specifying the metadata and linking it to the data itself. The generated tool includes the complete software stack, starting with a user interface, programmatic APIs for connecting additional application logic, and a persistence component. How these individual layers are realized is not specified, but defined by the mapping rules of the software generator, which also opens up the possibility of generating and evaluating different variants of the software.
Download
Paper Nr: 151
Title:

ML System Engineering Supported by a Body of Knowledge
Authors:

Juliette Mattioli, Dominique Tachet, Fabien Tschirhart, Henri Sohier, Loic Cantat and Boris Robert
Abstract: A body of knowledge (BoK) can be defined as the comprehensive set of concepts, terminology, standards, and activities that facilitate the dissemination of knowledge about a specific field, providing guidance for practice or work. This paper presents a methodology for the construction of a body of knowledge (BoK) based on knowledge-based artificial intelligence. The process begins with the identification of relevant documents and data, which are then used to capture concepts, standards, best practices, and state-of-the-art. These knowledge items are then fused into a knowledge graph, and finally, query capacities are provided. The overall process of knowledge collection, storage, and retrieval is implemented with the objective of supporting a trustworthy machine learning (ML) end-to-end engineering methodology, through the ML Engineering BoK.
Download
Paper Nr: 153
Title:

Developing an Artificial Intelligence Model to Enhance the Emotional Intelligence of Motor Vehicle Drivers for Safer Roads
Authors:

Ana Todorova, Irina Kostadinova and Svetlana Stefanova
Abstract: Traffic accidents and risky driving behaviour are among the deadliest problems worldwide. This statement becomes an undeniable fact, thanks to the grim statistics of the World Health Organization, according to which more than 1 million people die on the roads every year. Road accidents are also among the leading causes of death among children and young people aged 5 to 29. Against this background, a number of studies look for a link between the emotional intelligence of motor vehicle drivers and the potential prevention of risky driving. Building on the scientific knowledge generated up to this point, the present study suggests a prototype of an AI-based model that aims, through ongoing assessment and subsequent training, to enhance the emotional intelligence of both future and current motor vehicle drivers who are prone to risky behaviour on the road. Through simulated scenarios in a virtual environment, the model aims to improve the ability of drivers to recognise and manage their own and other people's emotions and to react adequately to different situations on the road. The expectation is that the model will reduce the manifestations of aggression and intolerance on the road and ultimately lead to safer roads.
Paper Nr: 185
Title:

Predicting Agricultural Product and Supplies Prices Using Artificial Intelligence
Authors:

Ioannis Dionissopoulos, Fotis Assimakopoulos, Dimitris Spiliotopoulos, Dionisis Margaris and Costas Vassilakis
Abstract: This work focuses on the prediction of agricultural product and supply prices using historical data and artificial intelligence methods. Agricultural product and supply prices are important for the economy and growth of agriculture. Using modern data analysis and deep learning methods, a forecasting model was developed to help us predict future price trends. The data used include the sales prices of crop products and the purchase prices of agricultural inputs. The developed forecasting methods exhibit high accuracy for predicting the actual prices of products and supplies, with error margins ranging from 0.29% to 9.8%, while they can also predict price rises and falls, with respective success rates ranging from 73.29% to 84.96%.
Download
Paper Nr: 55
Title:

Smart Data Stewardship: Innovating Governance and Quality with AI

Authors:

Otmane Azeroual
Abstract: In the modern digital landscape, data plays a crucial role in the competitiveness and efficiency of organizations. Data governance, which involves managing and ensuring data quality, faces increasing challenges due to the growing volumes and complexities of data. This paper examines how artificial intelligence (AI) offers innovative solutions for optimizing data governance and data quality. We present an AI-powered framework that includes components such as data integration, quality assurance, data protection monitoring, and compliance management. Through case studies and practical examples, we demonstrate how this framework can be implemented in real-world environments and the benefits it offers.
Download
Paper Nr: 58
Title:

EX-DSS: An Explorative Decision Support System for Designing and Deploying Smart Plug Forecasting Pipelines
Authors:

Giulia Rinaldi, Lola Botman, Oscar M. Agudelo and Bart De Moor
Abstract: Artificial Intelligence pipelines are increasingly used to address specific challenges, such as forecasting smart plug loads. Smart plugs, which remotely control various appliances, can significantly reduce energy consumption in commercial buildings by about 20% when effectively scheduled using AI techniques. Designing these AI pipelines involves numerous steps and variables, requiring collaboration and shared knowledge among designers. A Decision Support System (DSS) can facilitate this process. This paper introduces the Explorative Decision Support System (EX-DSS), which extends the classical DSS framework. The EX-DSS integrates an Explorative Management Subsystem to provide project-specific recommendations and a Data Quality (DQ) module to validate user inputs, ensuring clarity and enhancing information sharing. The EX-DSS architecture framework was tested through a software prototype designed to create AI pipelines for forecasting smart plug loads. The study found that using the EX-DSS improves the quality of suggestions, making them more problem-specific and resulting in a more personalized and meaningful user experience, with a significant potential to reduce energy consumption in commercial buildings.
Download
Paper Nr: 59
Title:

Assessing the Use of Online Platforms in Sharing Tacit Knowledge in Innovation Networks
Authors:

Nathalya Guruge and Jyri Vilko
Abstract: In an increasingly digitalized society, sharing tacit knowledge has emerged as a critical activity for driving innovation, especially within innovation networks. This paper presents a systematic literature review to assess the role of online platforms in facilitating tacit knowledge sharing. It explores how digital tools impact the exchange of tacit knowledge, offering a conceptual framework to understand this process. The findings provide strategies for leveraging online platforms to foster innovation within diverse knowledge-driven ecosystems.
Download
Paper Nr: 61
Title:

Development of a Concept Map Evaluation Support System for Social Studies Learning
Authors:

Keitaro Tokutake, Dai Sakuma and Masao Murota
Abstract: In social studies learning, it is crucial for students to develop a "structural awareness" that systematically organizes the connections between social phenomena. One approach to achieving this is concept mapping, and Tokutake et al. (2019) developed the S-R Score Table as a method for teachers to evaluate students' concept maps. However, the procedure for utilizing this method is complex, and interpreting the results requires specialized knowledge and insight. Therefore, in this study, we developed an evaluation support system that automates the creation of the S-R Score Table and displays the comparison results of the concept maps created by teachers and students in a comprehensive view. This system is designed to make it easier for teachers to evaluate the overall trends in students' structural awareness. The application of this system in actual classroom settings suggested that it could enhance teachers' ability to evaluate the structural awareness trends of the entire class.
Download
Paper Nr: 77
Title:

A Comprehensive Approach for Graph Data Warehouse Design: A Case Study for Learning Path Recommendation Based on Career Goals
Authors:

Viet Nguyen, Vu Bui and Thu Nguyen
Abstract: Today, organizations leverage big data analytics for insights and decision-making, handling vast amounts of structured and unstructured data. Traditional data warehouses (TDW) are suboptimal for such analytics, creating a demand for NoSQL-based modern data warehouses (DWs) that offer improved storage, scalability, and unstructured data processing. Graph-based data models (GDMs), a common NoSQL data model, are considered the next frontier in big data modeling. They organize complex data points based on relationships, enabling analysts to see connections between entities and draw new conclusions. This paper provides a comprehensive methodology for graph-based data warehouse (GDW) design, encompassing conceptual, logical, and physical phases. In the conceptual stage, we propose a high-abstraction data model for NoSQL DW, suitable for GDM and other NoSQL models. During the logical phase, GDM is used as the logical DW model, with a solution for mapping the conceptual DW model to GDW. We illustrate the GDW design phases with a use case for learning path recommendations based on career goals. Finally, we carried out the physical implementation of the logical DW model on the Neo4j platform to demonstrate its efficiency in managing complex queries and relationships, and showcase the applicability of the proposed model.
Download
Paper Nr: 87
Title:

Knowledge Management in Civil Protection at the Example of Fire Brigades
Authors:

Andreas Schultz, Fabian Dotzki and Iryna Mozgova
Abstract: Knowledge management is essential for successful disaster management. This paper conducts a Systematic Literature Review at the intersection of the knowledge management field and disaster management and examines the available body of literature. Fire departments are chosen as the focus group as they are the most prevalent emergency services. There are many publications that deal with knowledge management during the response phase of an emergency. Often, the literature focuses on the application of knowledge management in large-scale disasters to link the various organizations on-scene. What is missing in most approaches is a prior step of implementing and training the knowledge management system. Therefore, this literature review seeks to provide an overview of approaches for daily routines and small-to-medium incidents that serve as a training ground. However, literature on non-incident phases and smaller incidents is scarce. As information technologies are developing rapidly, there is no modern and recent description of the current use of knowledge management solutions in this area.
Download
Paper Nr: 109
Title:

Exploring Knowledge Sharing Motivational Factors for Intellectual Property Lawyers: A Conceptual Framework
Authors:

Nopphawat Ariyahchatraweekul, Mongkolchai Wiriyapinit and Jintavee Khlaisang
Abstract: This study aims to explore the motivational factors affecting knowledge sharing among lawyers in intellectual property law firms. Through an integrative review of existing empirical and theoretical literature, the research develops a conceptual framework to understand these motivational factors. The findings identify eight key extrinsic factors and two intrinsic factors influencing knowledge sharing for lawyers in intellectual property law firms. This research provides valuable insights into motivational factors that can be used by intellectual property law firms to improve knowledge sharing practices among lawyers.
Download
Paper Nr: 172
Title:

Connecting Critical Infrastructure Operators and Law Enforcement Agencies to Share Cyber Incident Information with Early Warning Systems
Authors:

Harri Ruoslahti and Ilkka Tikanmäki
Abstract: Cyber incidents and business interruptions rank as the foremost business risks. With Early Warning Systems (EWS), that work in parallel with other cyber mechanisms, organisations can independently manage cyber-sensitive intelligence-related data. This article provides a qualitative multi-case study analysis. The data consists of systematic reviews and cross-case conclusions of six (n = 6) case studies on information sharing. EWS is a valuable tool that can help critical infrastructure providers protect against cyberattacks. EWS can provide a platform for sharing information and resources. This can help improve situational awareness, enhance incident response, and facilitate collaboration. between critical infrastructure providers, as critical infrastructure operators and relevant Law Enforcement Agencies (LEA) can share information on cyber incidents and monitor cyber incident progress. EWS can be used to exchange cyber threat intelligence and information sharing can be facilitated with a common reference library where alerts can be shared as tickets. This would enable information exchange in both directions.
Download
Paper Nr: 173
Title:

Development Areas in Knowledge Management Processes in Social and Health Care Services: A Pilot Study
Authors:

Nina Helander, Annamaija Paunu, Pasi Hellsten, Hannele Väyrynen and Virpi Sillanpää
Abstract: Knowledge management is about providing the right information for the decision maker at the right time, in the right format. This is equally essential in public sector as in the private sector enterprises. However, it is often easier said than done where and how the development schemes are to be directed. More data and information is needed. Public sector’s social and health care has a number of data sources where knowledge management could make difference both in operational side, i. e. the care function and also in the management of the function, i. e. resource allocation. However, these are quite often planned some time ago and in the need of rethinking. The paper explores the possible points to be developed based on the knowledge management process, how to combine the two for a better outcome.
Download
Paper Nr: 184
Title:

Transforming Knowledge Management Using Generative AI: From Theory to Practice
Authors:

Dmitry Kudryavtsev, Umair A. Khan and Janne Kauttonen
Abstract: Generative AI is revolutionizing the way people and companies create, capture and access knowledge. This study is driven by problems and new opportunities related to knowledge work. We identify, organize, and prioritize generative AI use cases for knowledge management. Our analysis of business needs and in-depth interaction with companies is the main data source used to create insights into this study. In addition to the use cases, the research highlights the challenges of using generative AI for knowledge management and existing research tasks. Creating a reusable toolkit for Generative AI-enhanced knowledge management is proposed as the next step of applied research to address the identified use cases and challenges.
Download