Database and Expert Systems Applications - DEXA 2021 Workshops: BIOKDD, IWCFS, MLKgraphs, AI-CARES, ProTime, AISys 2021, Virtual Event, September ... in Computer and Information Science)

This volume constitutes the refereed proceedings of the workshops held at the 32nd International Conference on Database and Expert Systems Applications, DEXA 2021, held in a virtual format in September 2021: The 12 th International Workshop on Biological Knowledge Discovery from Data (BIOKDD 2021), the 5 th International Workshop on Cyber-Security and Functional Safety in Cyber-Physical Systems (IWCFS 2021), the 3 rd International Workshop on Machine Learning and Knowledge Graphs (MLKgraphs 2021), the 1 st International Workshop on Artificial Intelligence for Clean, Affordable and Reliable Energy Supply (AI-CARES 2021), the 1 st International Workshop on Time Ordered Data (ProTime2021), and the 1 st International Workshop on AI System Engineering: Math, Modelling and Software (AISys2021). Due to the COVID-19 pandemic the conference and workshops were held virtually. The 23 papers were thoroughly reviewed and selected from 50 submissions, and discuss a range of topics including: knowledge discovery, biological data, cyber security, cyber-physical system, machine learning, knowledge graphs, information retriever, data base, and artificial intelligence. Preface Organization Contents Cyber-Security and Functional Safety in Cyber-Physical Systems Mode Switching for Secure Web Applications – A Juice Shop Case Scenario 1 Introduction 2 Mode Switching 3 Juice Shop 4 Securing Juice Shop 5 Mode Switches in Juice Shop 6 Discussion 7 Conclusion References A Conceptual Model for Mitigation of Root Causes of Uncertainty in Cyber-Physical Systems 1 Introduction 2 Literature Review 3 Root Causes of Uncertainty in CPS 3.1 Human Behavior 3.2 Technological Process 3.3 Natural Process 4 Conceptual Model 5 Conclusion and Future Work References Security-Based Safety Hazard Analysis Using FMEA: A DAM Case Study 1 Introduction 2 Background 3 Case Study Design 4 Results 4.1 Security-Based Safety Hazard Analysis of DAM Case Study Using FMEA 4.2 Safety Requirements 5 Conclusion and Future Work References Privacy Preserving Machine Learning for Malicious URL Detection 1 Introduction 2 FHE and Privacy Preserving Machine Learning 3 Our Approach 3.1 DNN Approach 3.2 Logistic Regression Approach 3.3 Hybrid Approach 4 Conclusion and Future Work References Remote Attestation of Bare-Metal Microprocessor Software: A Formally Verified Security Monitor 1 Introduction 2 State of the Art 3 Extension to Microprocessors 3.1 Environment 3.2 Refinement Approach 4 Contribution 5 Verified Security Monitor 5.1 Attacker Model in Practice 5.2 Attesting Software and Monitor Architectures 5.3 Definition of Model 0 5.4 Proof Strategy 5.5 Results and Future Work 6 Conclusion References Provenance and Privacy in ProSA 1 Introduction 2 Provenance, Privacy, and Research Data Management 3 Preparation 4 Question Catalog 5 Evaluation 6 Conclusion References Machine Learning and Knowledge Graphs Placeholder Constraint Evaluation in Simulation Graphs 1 Introduction 2 Simulation Creation Process 2.1 Technical Aspects 2.2 Definition of Requirements and Provisions 3 Placeholder Constraints Evaluation 4 Summary References Walk Extraction Strategies for Node Embeddings with RDF2Vec in Knowledge Graphs 1 Introduction 2 Background 2.1 Weisfeiler-Lehman Kernel for Knowledge Graphs 3 Custom Walk Extraction Strategies 4 Results 4.1 Datasets 4.2 Setup 4.3 Evaluation Results 5 Discussion 6 Conclusion and Future Work References Bridging Semantic Web and Machine Learning: First Results of a Systematic Mapping Study 1 Introduction 2 Survey Methodology: Systematic Mapping Study 2.1 Study Planning 2.2 Study Execution 3 Initial Results 4 Conclusion and Outlook References On the Quality of Compositional Prediction for Prospective Analytics on Graphs 1 Introduction 2 Data Model and Motivation Scenario: Rennes City Bus Transportation 3 Micro-learning and Compositional Prediction 4 Experiments 4.1 Experimental Settings 4.2 Results 4.3 Discussion 4.4 Threats to Validity 5 Related Work 5.1 Micro Learning 5.2 Link Weight Prediction 5.3 Bus Travel Time/Bus Speed Prediction 5.4 Traffic Simulation 6 Conclusion and Future Work References Semantic Influence Score: Tracing Beautiful Minds Through Knowledge Diffusion and Derivative Works 1 Introduction 2 Related Works 3 Data Description 4 Methodology 4.1 Filtering Data-Set 4.2 Tagging Documents into Categories 4.3 Creating Semantic Influence Network (SIN) of Influential Papers and Their Derivatives 4.4 Influence Score 4.5 Similarity Score 4.6 Semantic Influence Score (SIS) 5 Results 6 Discussion and Conclusion References AI System Engineering: Math, Modelling and Software Robust and Efficient Bio-Inspired Data-Sampling Prototype for Time-Series Analysis 1 Introduction 1.1 Weyl's Discrepancy 1.2 Adaptive Threshold-Based Sampling Approach Based on Quasi-Isometry 2 Prototype Development 2.1 Spike Data Representation and Communication 3 Measurement Results 4 ECG Measurements 5 Conclusion References Membership-Mappings for Data Representation Learning: Measure Theoretic Conceptualization 1 Introduction 2 Notations and Definitions 3 Representation of Samples via Attribute Values 3.1 A Measure Space 3.2 Student-t Membership-Mapping 3.3 Interpolation by Student-t Membership-Mapping 3.4 Variational Learning of Membership-Mappings 3.5 Prediction by Membership-Mappings 4 Concluding Remarks References Membership-Mappings for Data Representation Learning: A Bregman Divergence Based Conditionally Deep Autoencoder 1 Introduction 2 A Bregman Divergence Based Conditionally Deep Autoencoder 2.1 Variational Learning Algorithm 2.2 A Wide Conditionally Deep Autoencoder 3 Classification Applications 4 Experiments 4.1 Classification of High-Dimensional Image Features 4.2 Robustness in Classification 5 Concluding Remarks References Data Catalogs: A Systematic Literature Review and Guidelines to Implementation 1 Introduction 2 Research Method 2.1 Research Questions 2.2 Search Strategy 2.3 Paper Selection Process 3 Results from the Literature Review 3.1 Overview on Data Catalog Implementations in Practice 3.2 Components of a Data Catalog 3.3 Guidelines to Implement a Data Catalog 4 Discussion and Outlook References Task-Specific Automation in Deep Learning Processes 1 Introduction 1.1 Related Work 1.2 Machine Learning Pipelines 1.3 Automatic ML 1.4 Task Specific Pipelines 2 ALOHA Tool Flow 3 Evaluation and Results 3.1 Surveillance Use Case 3.2 Smart Industry Use Case 3.3 Medical Use Case 3.4 Evaluation Method 3.5 Preliminary Results 4 Conclusion References Time Ordered Data Approximate Fault Tolerance for Edge Stream Processing 1 Introduction 2 Related Work 3 Preliminaries 4 Proposed Approach 4.1 Overview 4.2 Aggregation Result Estimation 5 Experimental Evaluation 5.1 Experimental Settings 5.2 Methodology 5.3 Results 6 Conclusions References Deep Learning Rule for Efficient Changepoint Detection in the Presence of Non-Linear Trends 1 Introduction 2 The Problem: Anomalies in the Presence of Trends 3 Description of the Experiment 4 Discussion 4.1 Application to Synthetic Data 4.2 Empirical Study: Application to Food Production Data 5 Conclusions References Time Series Pattern Discovery by Deep Learning and Graph Mining 1 Introduction 2 Related Work 3 Methods 3.1 Transform Time Series Data to Embedded Vectors 3.2 Transform Vectors to Images 3.3 Train CNN Image Classification Model 3.4 Graph Mining of Time Series Data 4 Experiments 4.1 Data Source 4.2 Data Transformation: From Raw Data to Embedding Space 4.3 Data Transformation: From Vectors to Images 4.4 CNN Image Classification 4.5 EEG Channel Positions 4.6 Build Time Series Graph and Find Hidden Patterns 4.7 Graph Cluster Illustration 5 Conclusion 6 Broader Impact References Biological Knowledge Discovery from Big Data Integrating Gene Ontology Based Grouping and Ranking into the Machine Learning Algorithm for Gene Expression Data Analysis 1 Introduction 2 Materials and Methods 2.1 Gene Expression Data 2.2 Gene Ontology Data 3 Methods 3.1 Algorithm 3.2 Implementation 3.3 Model Performance Evaluation 4 Results and Discussions 5 Conclusion References SVM-RCE-R-OPT: Optimization of Scoring Function for SVM-RCE-R 1 Introduction 2 Methods and Implementation 2.1 SVM-RCE 2.2 SVM-RCE-R 2.3 SVM-RCE-R Optimal 2.4 KNIME Workflow 3 Data 4 Results 5 Conclusions References Artificial Intelligence for Clean, Affordable and Reliable Energy Supply Short-Term Renewable Energy Forecasting in Greece Using Prophet Decomposition and Tree-Based Ensembles 1 Introduction 2 Related Work 3 Dataset Description 4 Forecasting Models 4.1 Baseline Models 4.2 Machine Learning Models 4.3 Proposed Hybrid Model 5 Experimental Results 6 Conclusions and Future Work References A Comparative Study of Deep Learning Approaches for Day-Ahead Load Forecasting of an Electric Car Fleet 1 Introduction 2 Deep Learning Models 3 Data Pre-processing and Feature Analysis 4 Experimental Results and Discussion 5 Conclusions and Future Work References Correction to: A Comparative Study of Deep Learning Approaches for Day-Ahead Load Forecasting of an Electric Car Fleet Correction to: Chapter “A Comparative Study of Deep Learning Approaches for Day-Ahead Load Forecasting of an Electric Car Fleet” in: G. Kotsis et al. (Eds.): Database and Expert Systems Applications - DEXA 2021 Workshops, CCIS 1479, https://doi.org/10.1007/978-3-030-87101-7_23 Author Index