Big Data Analytics and Knowledge Discovery: 24th International Conference, DaWaK 2022, Vienna, Austria, August 22–24, 2022, Proceedings (Lecture Notes in Computer Science, 13428)

This volume LNCS 13428 constitutes the papers of the 24 th International Conference on Big Data Analytics and Knowledge Discovery, held in August 2022 in Vienna, Austria. The 12 full papers presented together with 12 short papers in this volume were carefully reviewed and selected from a total of 57 submissions. The papers reflect a wide range of topics in the field of data integration, data warehousing, data analytics, and recently big data analytics, in a broad sense. The main objectives of this event are to explore, disseminate, and exchange knowledge in these fields. Preface Organization Contents Text Analytics An Integration of TextGCN and Autoencoder into Aspect-Based Sentiment Analysis 1 Introduction 2 Methodology 2.1 Problem Description 2.2 Aspect Extraction 2.3 Aspect-Based Sentiment Words Generation 2.4 Aspect Sentiment Detection 3 Experiments 3.1 Dataset 3.2 Experimental Settings 3.3 Evaluation Metrics 3.4 Experimental Results on Aspect Extraction 3.5 Experimental Results on Aspect-Sentiment Identification 3.6 Discussion of Aspect-Based Sentiment Words 4 Conclusion References OpBerg: Discovering Causal Sentences Using Optimal Alignments 1 Introduction 2 Preliminaries 2.1 Sequence Alignments 2.2 AGE 2.3 Causality in Biological Literature 2.4 Text Search and Extraction 3 Proposed Approach 3.1 OpBerg 4 Results 4.1 Classification 5 Discussion 6 Conclusion References Text-Based Causal Inference on Irony and Sarcasm Detection 1 Introduction 2 Related Works 3 Background 3.1 Causal Inference 3.2 NLP with Causality 3.3 Causal Model Explainability 4 Methods 4.1 Text-Based Causal Inference Using TextCause 4.2 Unsupervised Data Analysis for Determining Confounders 4.3 Modeling Causal Inference for Irony and Sarcasm Detection 5 Experiments 5.1 Dataset and Settings 5.2 Results 6 Conclusion References Sarcastic RoBERTa: A RoBERTa-Based Deep Neural Network Detecting Sarcasm on Twitter 1 Introduction 2 Datasets 3 Proposed Approach 4 Evaluation 5 Related Work 6 Discussion and Conclusions References A Fast NDFA-Based Approach to Approximate Pattern-Matching for Plagiarism Detection in Blockchain-Driven NFTs 1 Introduction 2 Related Work 3 Implementation 4 Experimental Results 5 Conclusions References Data Warehousing and OLAP On Decisive Skyline Queries 1 Introduction 2 Problem Formulation 2.1 Intuition of Decisive Subspaces 2.2 Formal Definition of Decisive Subspaces 2.3 Decisive Skyline Points 3 Decisive Skyline Algorithm 3.1 Algorithmic Description 4 Experimental Evaluation 4.1 Qualitative Study 4.2 Performance of Decisive Skyline Algorithm 4.3 Comparison with Representative Skylines 5 Conclusions References Safeness: Suffix Arrays Driven Materialized View Selection Framework for Large-Scale Workloads 1 Introduction 2 Related Work 3 Suffix Arrays for Large-Scale Workload Coding 3.1 Fundamental Concepts of Suffix Arrays 3.2 Application of Suffix Arrays to OLAP Queries 4 Safeness Framework for MV Selection 5 Experimental Study 6 Conclusion References A Process Warehouse for Process Variants Analysis 1 Introduction 2 Generate a Process Variant Hierarchy 3 Related Work 4 Conclusions References Feature Selection Algorithms Unsupervised Features Ranking via Coalitional Game Theory for Categorical Data 1 Introduction 2 Related Work 3 Feature Importance Measures 3.1 Feature Importance Score 3.2 Importance Scores of Low Correlated Features 4 Redundancy Removal 5 Scalable Algorithms 6 Experiments 6.1 Datasets and Competing Methods 6.2 Redundancy Awareness 6.3 Relevance of Unsupervised Feature Selection and Effectiveness 6.4 Interpretation of Feature Ranking 6.5 Comparison Among the Proposed Algorithms 6.6 Run-Time Analysis 7 Conclusions References Multi-label Online Streaming Feature Selection Algorithms via Extending Alpha-Investing Strategy 1 Introduction 2 Multi-label Online Streaming Feature Selection Algorithms via Extending Alpha-Investing Strategy 2.1 Preliminaries 2.2 Alpha-Investing Method for Single-Label OSFS 2.3 Multi-label Online Streaming Feature Selection via Combining Alpha-Investing with Binary Relevance 2.4 Multi-label Online Streaming Feature Selection via Combining Alpha-Investing with Multi-output Regression 3 Experiments 4 Conclusions References Feature Selection Under Fairness and Performance Constraints 1 Introduction 2 Related Work 3 The Proposed Method 3.1 Algorithm of the Proposed Method 4 Experimental Setup 4.1 Results Analysis 5 Conclusion References Time Series Processing Interpretable Input-Output Hidden Markov Model-Based Deep Reinforcement Learning for the Predictive Maintenance of Turbofan Engines 1 Introduction 2 Related Work 3 Framing Predictive Maintenance as an RL Problem 3.1 Environment Dynamics and Modeling 3.2 Reward Formulation 3.3 Evaluation Criteria 4 Proposed Methodology (SRLA) 4.1 Interpretability with IOHMM 5 Experimental Setup 5.1 Setting the Hyperparameters for the Models 6 Experiment 1: Interpretations Based on Hidden States 6.1 Interpretability - Failure Event Hypothesis 6.2 Interpretability - State Decoding and Mapping 7 Experiment 2: Comparison of SRLA with Prior Work 7.1 Comparative Evaluation and Results 8 Conclusion and Future Direction A Algorithms and Training Parameters A.1 Training Parameters References Pathology Data Prioritisation: A Study Using Multi-variate Time Series 1 Introduction 2 Previous Work 3 U&E Testing Application Domain 4 Prioritisation for U&E Pathology Patients Results 4.1 Future Results Prediction 4.2 Bounding Box Classification 5 Evaluation 5.1 Evaluation Data Set 5.2 Data Imputation 5.3 Overall Performance 6 Conclusions References Outlier/Anomaly Detection of Univariate Time Series: A Dataset Collection and Benchmark 1 Introduction 2 Related Work 3 Dataset Collection 3.1 Detection Tasks 3.2 Detection Benchmarks 3.3 Evaluation and Baselines 4 Conclusions References Schema Discovery and Construction Automatic Machine Learning-Based OLAP Measure Detection for Tabular Data 1 Introduction 2 Related Works 3 Measure Detection 3.1 Overview 3.2 Preprocessing 3.3 Feature Extraction 4 Experimental Validation and Discussion 4.1 Experimental Conditions 4.2 Baseline Methods 4.3 Experimental Results 5 Conclusion and Future Work References Discovering Overlapping Communities Based on Cohesive Subgraph Models over Graph Data 1 Introduction 2 Formal Preliminaries 3 Novel Cohesive Models for Community Discovery 4 Experimental Evaluation 4.1 Choosing the Best Value of Parameters 4.2 Comparison with Existing Cohesive Models 4.3 Comparison with Baseline Algorithms 5 Conclusion References Discovery of Keys for Graphs 1 Introduction 2 Preliminaries 3 Discovering Graph Keys 3.1 Key Properties 3.2 GKMiner Algorithm 4 Experiments 5 Conclusion and Future Work References OPTIMA: Framework Selecting Optimal Virtual Model to Query Large Heterogeneous Data 1 Introduction 2 OPTIMA: Optimal Virtual Model for Querying Large Heterogeneous Data 2.1 Virtual Data Model Prediction 2.2 Query Decomposition and Relevant Entity Detection 2.3 Data Wrapper 2.4 Distributed Query Processor 3 Experimental Setup 4 Related Work 5 Conclusion References Pattern Discovery Q-VIPER: Quantitative Vertical Bitwise Algorithm to Mine Frequent Patterns 1 Introduction 2 Background and Related Works 2.1 Vertical Boolean Frequent Pattern Mining with the VIPER Algorithm 2.2 Quantitative Association Rule Mining 2.3 Horizontal Quantitative Frequent Pattern Mining with the MQA-M Algorithm 3 Vertical Quantitative Frequent Pattern Mining with Our Q-VIPER Algorithm 3.1 Vertical Representation of Quantitative Data 3.2 Q-VIPER Algorithm 3.3 Our New Pruning Rules for Q-VIPER Algorithm 4 Evaluation 5 Conclusions References Enhanced Sliding Window-Based Periodic Pattern Mining from Dynamic Streams 1 Introduction 2 Background and Related Works 3 Our Sliding Window-Based Weighted Periodic Pattern Mining Algorithm 4 Evaluation 5 Conclusions References Explainable Recommendations for Wearable Sensor Data 1 Introduction 2 Background and Motivation 3 Mining Frequent Patterns for Recommending Activities 4 Conclusions References Machine Learning SLA-Aware Cloud Query Processing with Reinforcement Learning-Based Multi-objective Re-optimization 1 Introduction 2 The Reinforcement Learning-Based Multi-objective Query Re-optimization Algorithm (ReOptRL) 3 The SLA-Aware Reinforcement Learning-Based Multi-objective Query Re-optimization Algorithm (SLAReOptRL) 4 Performance Evaluation 5 Conclusion References Mahalanobis Distance Based K-Means Clustering 1 Introduction 2 Background and Related Work 3 Our Clustering Algorithm 4 Evaluation 5 Conclusions References Grapevine Phenology Prediction: A Comparison of Physical and Machine Learning Models 1 Introduction 2 Background of Phenology and Prediction Models 3 Experimental Set up and Evaluation 3.1 Raw Data Pre-processing and Analysis 3.2 Experimental Results of the Physical Models 3.3 Experimental Results of the ML Models 4 Conclusions and Future Work References Author Index