Knowledge Discovery, Knowledge Engineering and Knowledge Management : 14th International Joint Conference, IC3K 2022, Valletta, Malta, October 24-26, 2022, Revised Selected Papers / / Frans Coenen [and six others], editors

• Record Nr.: UNISA996550556503316
• Pubbl/distr/stampa: Berlin, Germany : , : Springer, , [2023]; ©2023
• ISBN: 3-031-43471-4
• Edizione: [First edition.]
• Descrizione fisica: 1 online resource (368 pages)
• Collana: Communications in Computer and Information Science Series ; ; Volume 1842
• Disciplina: 006.3
• Soggetti: Data mining; Information retrieval; Knowledge management
• Lingua di pubblicazione: Inglese
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Nota di bibliografia: Includes bibliographical references and index.
• Nota di contenuto: Intro -- Preface -- Organization -- Contents -- Knowledge Discovery and Information Retrieval -- Electrocardiogram Two-Dimensional Motifs: A Study Directed at Cardio Vascular Disease Classification -- 1 Introduction -- 2 Related Work -- 3 Problem Definition -- 4 Cardiovascular Disease Classification Model Generation -- 4.1 Data Cleaning (Stage 1) -- 4.2 Motif and Discord Extraction (Stage 2) -- 4.3 Feature Selection (Stage 3) -- 4.4 Data Augmentation (Stage 4) -- 4.5 Feature Vector Generation (Stage 5) -- 4.6 Classification Model Generation (Stage 6) -- 4.7 Classification Model Usage (Stage 7) -- 5 Evaluation -- 5.1 Data Sets -- 5.2 Most Appropriate Feature Selection and Data Augmentation Techniques (Objective 1) -- 5.3 Most Appropriate Conflict Resolution Technique (Objective 2) -- 5.4 Operation Using Additional Features (Objective 3) -- 5.5 Comparison of 1D and 2D Motifs Discovery Approaches (Objective 4) -- 6 Conclusion -- References -- Degree Centrality Definition, and Its Computation for Homogeneous Multilayer Networks Using Heuristics-Based Algorithms -- 1 Motivation -- 1.1 Differences with the KDIR Conference Paper -- 2 Relevant Work -- 3 Decoupling Approach for Multilayer Networks -- 4 ; Degree Centrality for Graphs and Homogeneous MLNs -- 4.1 Impact of Layer Information on Accuracy -- 5 Accuracy of Degree Centrality Heuristics -- 5.1 First Heuristic for Accuracy (DC-A1) -- 5.2 Second Heuristic for Accuracy (DC-A2) -- 6 Heuristics for Precision -- 6.1 Heuristic 1 for Precision (DC-P1) -- 6.2 Heuristic 2 for Precision (DC-P2) -- 7 Data Sets and Computation Environments -- 7.1 Data Sets -- 7.2 Computation Environments Used -- 8 Discussion of Experimental Results -- 9 Conclusions and Future Work -- References -- A Dual-Stage Noise Training Scheme for Breast Ultrasound Image Classification -- 1 Introduction.; 1.1 Image Data Predicament in Medical Area -- 1.2 Related Work -- 1.3 Contributions -- 2 Speckle Noise -- 3 Methodology -- 3.1 Dataset Preparation -- 3.2 CNN Model Selection -- 3.3 Performance Metrics -- 3.4 A Dual-Stage Noise Training Scheme -- 4 Experiment Results -- 4.1 Stage 1 -- 4.2 Stage 2 -- 5 Conclusions -- References -- A General-Purpose Multi-stage Multi-group Machine Learning Framework for Knowledge Discovery and Decision Support -- 1 Introduction -- 2 Optimization-Based Classification Models -- 2.1 A Multi-group Machine Learning Framework -- 2.2 A Multi-stage Multi-group Machine Learning Framework -- 2.3 Balancing Misclassification Levels vs Size of the Reserve Judgement Region -- 2.4 Applying Multi-stage BB-PSO/DAMIP to Real-World Problems -- 3 Results for Disease Diagnosis and Treatment Outcome prediction -- 3.1 Cardiovascular Disease -- 3.2 Diabetes -- 3.3 Alzheimer's Disease -- 3.4 Knee Osteoarthritis -- 4 Discussions -- References -- Comparative Assessment of Deep End-To-End, Deep Hybrid and Deep Ensemble Learning Architectures for Breast Cancer Histological Classification -- 1 Introduction -- 2 Material and Methods -- 2.1 Deep Learning and Transfer Learning -- 2.2 Experiment Configuration and Design -- 2.3 Data Preparation -- 2.4 Abbreviation -- 3 Results and Discussions -- 3.1 Overall Performance of the Deep End-to-End Architectures -- 3.2 Performance Comparison of Deep end-to-end Architectures -- 3.3 Comparison of Deep end-to-end, Hybrid and End-to-end Ensemble Learning Architectures -- 4 Threats of Validity -- 5 Conclusion and Future Work -- Appendix A: Deep Architectures four Performance Measures Validation Results -- References -- Knowledge Engineering and Ontology Development -- CIE: A Cloud-Based Information Extraction System for Named Entity Recognition in AWS, Azure, and Medical Domain -- 1 Introduction and Motivation.; 2 State of the Art in Science and Technology -- 2.1 Named Entity Recognition -- 2.2 ML and Deep Learning in Named Entity Recognition -- 2.3 Cloud Resource Management for Named Entity Recognition -- 2.4 Named Entity Recognition Frameworks -- 2.5 Related Research Projects -- 3 CIE Modeling and Implementation -- 3.1 CIE AWS Implementation -- 3.2 CIE Azure Implementation -- 4 Final Discussion and Conclusion -- References -- From Natural Language Texts to RDF Triples: A Novel Approach to Generating e-Commerce Knowledge Graphs -- 1 Introduction -- 2 Related Work -- 3 Framework QART -- 3.1 Step A: Field Selection and Pre-processing -- 3.2 Step B: Text2Text Conversion -- 3.3 Step C: Text Triplifying -- 3.4 Implementation Aspects -- 4 Evaluating Template-Based Text Summarization -- 4.1 Setup and Procedures -- 4.2 Results -- 4.3 Discussion -- 5 Evaluating Automatic Text-to-Text Transformation -- 5.1 Setup and Procedures -- 5.2 Results -- 5.3 Discussion -- 6 Overall Discussion and Challenges -- 7 Conclusion -- References -- Situational Question Answering over Commonsense Knowledge Using Memory Nets -- 1 Introduction -- 2 Related Work -- 3 System Overview -- 3.1 Knowledge Engine -- 3.2 Semantic Parsing -- 3.3 Knowledge Extraction -- 3.4 ; Virtual Simulation -- 3.5 XAI -- 4 Evaluation -- 4.1 Instance Question Answering -- 4.2 Action Pattern Question Answering -- 5 Conclusion -- References -- Archives Metadata Text Information Extraction into CIDOC-CRM -- 1 Introduction -- 2 Related Work -- 3 Overview of Archives Metadata Representation in CIDOC-CRM -- 3.1 Extracted Information from ISAD(G) Elements -- 3.2 CIDOC-CRM Representation of the Events and Entities Extracted -- 4 The Extraction of Events and Entities from Semi-structured Text -- 4.1 Semantic Role Labelling Process Using GATE -- 4.2 Semantic Role Token Labelling Process Using BERT.; 5 Evaluation of the Extraction Process -- 5.1 Dataset -- 5.2 Evaluation Methodology and Results -- 5.3 Evaluation of the ANNIE Extraction Process -- 5.4 Evaluation of the BERT Extraction Process -- 6 Exploration of the Extracted Information -- 7 Conclusions and Future Work -- References -- Evolution of Computational Ontologies: Assessing Development Processes Using Metrics -- 1 Introduction -- 2 Related Work -- 3 Hypotheses on Ontology Evolution -- 4 Dataset Preparation and Analysis -- 5 Empirical Assessment of Hypotheses -- 5.1 Ontologies Grow During Their Lifetime (H1) -- 5.2 The Level of Change Decreases over Time (H2) -- 5.3 The Instances Are Introduced after the Initial Design (H3) -- 5.4 Ontology Complexity Increases with Rising Maturity (H4) -- 5.5 A Stereotypical Development Lifecycle Can Be Identified (H5) -- 6 Ontology Evolution or Revolution? -- 6.1 Most Ontologies Have Disruptive Change Events -- 6.2 The Size of Disruptive Change Events Varies -- 6.3 Disruptive Changes Come in Various Combinations -- 6.4 Sensitivity Analysis -- 7 Conclusion -- References -- System to Correct Toxic Expression with BERT and to Determine the Effect of the Attention Value -- 1 Introduction -- 2 Related Works -- 3 Proposed Method -- 3.1 Collecting Tweets -- 3.2 Preprocessing of Tweets -- 3.3 Creating a BERT Classifier -- 3.4 MASK Processing Conversion with BERT -- 3.5 Similarity Evaluation -- 4 Experimental Results -- 4.1 Classification Accuracy -- 4.2 Comparison of Various Patterns -- 4.3 Results of MASK Conversion by BERT -- 4.4 Results of the Three Evaluations -- 5 Conclusions -- References -- Knowledge Management and Information Systems -- Machine Learning Decision Support for Production Planning and Control Based on Simulation-Generated Data -- 1 Introduction -- 2 State of the Art -- 2.1 Production Planning and Control Systems.; 2.2 PPC Challenges and Possible Solutions -- 2.3 Fundamentals: Machine Learning -- 2.4 Related Work: Application of Machine Learning Within PPC -- 3 Framework for Development of an ML Decision Support System Based on Simulation Data -- 4 Case Study -- 4.1 Case Study Description -- 4.2 Results of the Case Study -- 5 Limitations -- 6 Conclusion and Outlook -- References -- FAIRification of CRIS: A Review -- 1 Introduction -- 2 Methodology -- 3 Results -- 3.1 FAIRification of CRIS -- 3.2 FAIRification of Workflows and Other Infrastructures -- 3.3 CRIS as an Input for RDM FAIRness Assessment -- 4 Discussion -- 4.1 Assessing the FAIRness of CRIS -- 4.2 Ecosystem -- 4.3 Factors for further FAIRification of CRIS -- 5 Conclusion -- Appendix 1: FAIR Principles -- Appendix 2 - Review Criteria -- References -- Measuring Augmented Reality and Virtual Reality Trajectory in the Training Environment -- 1 Introduction -- 2 Understanding the Terms -- 2.1 Defining Augmented Reality and Virtual Reality -- 2.2 History of Augmented Reality and Virtual Reality -- 2.3 Peer-Reviewed Research of Augmented and Virtual Reality -- 2.4 Virtual Reality in the Aviation Industry -- 3 Methodology in Measuring Effectiveness of VR Technology -- 4 Results of VR Technology versus Real-Life Simulation -- 4.1 The Timing of VR ; Technology Simulation in the Training Environment -- 4.2 VR Technology Simulation Resulting in Behavior Change -- 4.3 Findings in Non-VR Participants -- 5 Next Steps -- 6 Conclusion -- References -- DroNit Project: Improving Drone Usage for Civil Defense Applications -- 1 Introduction -- 2 Drone Types and Characteristics -- 3 Understanding the Needs of the Civil Defense of Niterói -- 3.1 Daily Activities and Roles of the Office -- 3.2 Current Use of Drones -- 3.3 Demands of Drone Usage -- 4 Challenges and Insights -- 4.1 Issues and Challenges.; 4.2 Recommendations for More Effective Drone Usage.