Web Information Systems Engineering – WISE 2021 : 22nd International Conference on Web Information Systems Engineering, WISE 2021, Melbourne, VIC, Australia, October 26–29, 2021, Proceedings, Part I

This two-volume set constitutes the proceedings of the 22nd International Conference on Web Information Systems Engineering, WISE 2021, held in Melbourne, VIC, Australia, in October 2021. The 55 full, 29 short and 5 demo papers, plus 2 tutorials were carefully reviewed and selected from 229 submissions. The papers are organized in the following topical sections: Part I: BlockChain and Crowdsourcing; Database System and Workflow; Data Mining and Applications; Knowledge Graph and Entity Linking; Graph Neural Network; Graph Query; Social Network; Spatial and Temporal Data Analysis. Part II: Deep Learning (1), Deep Learning (2), Recommender Systems (1), Recommender Systems (2), Text Mining (1), Text Mining (2), Service Computing and Cloud Computing (1), Service Computing and Cloud Computing (2), Tutorial and Demo. Preface Organization Contents – Part I Contents – Part II BlockChain and Crowdsourcing Crowdsourcing Software Vulnerability Discovery: Models, Dimensions, and Directions 1 Introduction 2 Models for Crowdsourcing Vulnerability Discovery 2.1 Direct Vulnerability Discovery 2.2 Platform Managed Vulnerability Discovery 2.3 Cyber Security Contests 3 Dimensions of Crowdsourcing Vulnerability Discovery 3.1 Crowd Size 3.2 Incentives 3.3 Duration of the Task 3.4 Task Context 3.5 Task Management 3.6 Selecting Security Professionals 3.7 Information Protection 3.8 Legal Terms 4 Discussion and Future Research Directions 4.1 Improving the Quality of Vulnerability Tasks Descriptions and Reports 4.2 Protecting Against Intellectual Property Leakage 4.3 Crowdsourcing Vulnerability Discovery Quality Analytics 5 Conclusion References Expertise-Aware Crowdsourcing Taxonomy Enrichment 1 Introduction 2 Problem Formulation 3 Our Method 3.1 Uncertain Instances Finding and Skill Matching 3.2 Subtree Recommendation 3.3 Skill Estimation 3.4 Truth Inference 4 Experiments 4.1 Baseline Methods 4.2 Evaluation Metrics 4.3 Simulated Experiment 4.4 Real-World Experiment 5 Related Works 6 Conclusion and Future Work References Transaction Confirmation Time Estimation in the Bitcoin Blockchain 1 Introduction 2 Background and Related Work 2.1 Blockchain Management 2.2 Transaction Confirmation in the Bitcoin Blockchain 2.3 Related Works on Transaction Confirmation Time Estimation 3 Problem Definition 4 Methods for Confirmation Time Estimation 4.1 Decayed Mean (DcyMean) 4.2 Confirmation Time Estimation Network (CTEN) 5 Experiments 5.1 Experiment Settings 5.2 Result Analysis 6 Conclusion References Automatic Malicious Worker Detection in Crowdsourced Paraphrases 1 Introduction 2 Cheating Behaviors 3 Cheating Detection 3.1 Feature Engineering 3.2 Malicious Worker Detection 4 Experiment and Results 4.1 Evaluation 4.2 Error Analysis 5 Related Work 6 Conclusion References A Blockchain-Based Approach for Trust Management in Collaborative Business Processes 1 Introduction 2 Related Work 3 Motivating Example 4 Methodology 4.1 Metrics to Support On-Chain Elements Selection 4.2 Abstract Smart Contracts Generation 5 Implementation and Validation Issues 6 Concluding Remarks and Future Research References Database System and Workflow Exploiting Unblocking Checkpoint for Fault-Tolerance in Pregel-Like Systems 1 Introduction 2 Background of Pregel-Like Systems 2.1 Execution Workflow 2.2 Checkpointing 3 Queuing Strategy 3.1 Resource Contention 3.2 Checkpoint Queuing 4 Skipping Policy 4.1 Checkpoint Staleness 4.2 Checkpoint Tardiness 4.3 Staleness/Tardiness-Aware Skipping 5 Experimental Studies 5.1 Experimental Setting 5.2 Efficiency of Queuing Strategy 5.3 Efficiency of Skipping Policy 6 Related Work 7 Conclusions References A Low-Latency Metadata Service for Geo-Distributed File Systems 1 Introduction 2 The Geo-Distributed File System Framework 3 A Low-Latency Metadata Service 4 Experiment Evaluation 4.1 Experiment Settings 4.2 Experiment Comparison Analysis 4.3 Evaluation Result 5 Related Works 6 Conclusion References XTuning: Expert Database Tuning System Based on Reinforcement Learning 1 Introduction 2 Related Work 3 Expert Knowledge-Based Tuning Architecture 3.1 Correlation Rules Table of Knobs 3.2 Knobs Correlation with Internal Expert Knowledge 3.3 Workloads Correlation with External Expert Knowledge 3.4 Progressive Expert Knowledge Tuning Algorithm 4 Experiment Study 4.1 Training Time Reduction with Expert Rules 4.2 Throughput Improvement 4.3 Latency Reduction 4.4 Architectural Optimization Performance in XTuning 5 Conclusion References CELA: An Accurate Learned Cardinality Estimator with Strong Generalization Ability and Dimensional Adaptability 1 Introduction 2 Related Work 3 Vectorization 4 Model 4.1 Considerations 4.2 Dynamic Vectorization 4.3 Dynamic Architecture 5 Experiment 5.1 Generating Data 5.2 Model Training 5.3 Evaluation 6 Conclusions References Cost-Based Lightweight Storage Automatic Decision for In-Database Machine Learning 1 Introduction 2 System Overview 3 Data Preparation 3.1 Data Partition 3.2 Feature Selection 3.3 Data Collection 4 Storage Decision Model 5 Experiments 5.1 Accuracy Evaluation 5.2 Feature Section Effectiveness Evaluation 5.3 Compare the Applicable Workloads of Row/Column Model 5.4 Comparisons on Various Workloads 6 Conclusions References Data Mining and Applications NP-PROV: Neural Processes with Position-Relevant-Only Variances 1 Introduction 2 Background 3 Neural Processes with Position-Relevant-Only Variances 3.1 Off-the-Grid Scenario 3.2 On-the-Grid Scenario 4 Experiments 4.1 Off-the-Grid Datasets 4.2 On-the-Grid Datasets 5 Related Work 6 Conclusions and Discussions A Compuational Complexity B Off-the-Grid Datasets Experiments C On-the-Grid Datasets Experiments References A Minority Class Boosted Framework for Adaptive Access Control Decision-Making 1 Introduction 2 Related Work 3 Methodology 3.1 Workflow of the Proposed Framework 3.2 Boosting Window Algorithm 4 Experiment Results 4.1 Dataset 4.2 Evaluation Metrics 4.3 Experimental Setting 4.4 Performance of Boosting Misclassified Samples 4.5 Performance Comparison with Different Negative Sample Rates 4.6 Discussion 5 Conclusion References Recognizing Hand Gesture in Still Infrared Images by CapsNet 1 Introduction 2 Related Work 2.1 CapsNet 2.2 Hand Gesture Recognition 3 Proposed Approach 3.1 Capsule Module 3.2 Reconstruction Module 4 Experimental Settings 4.1 Dataset 4.2 Dataset Split Modes 4.3 Training Parameters of the Proposed Approach 5 Experimental Results and Analysis 5.1 Experimental Results of Our Approach 5.2 Comparison of Confusion Matrix of Different Dataset Split Modes 5.3 Comparison with Existing Approaches 5.4 Comparison of Training Curve and Validation Curve for Different Dataset Split Modes 6 Ablation Study 6.1 Evaluation on Different Networks 6.2 Evaluation on Different Routing Iterations 6.3 Discussion 7 Conclusion and Future Work References Vertical Federated Principal Component Analysis on Feature-Wise Distributed Data 1 Introduction 2 Notations and Background 2.1 Notation 2.2 PCA and Power Iteration 2.3 Federated Learning 3 Federated-PCA on Privacy-Preserving Vertical-Partitioned Data 3.1 Problem Formulation 3.2 Local Power Method 3.3 Federated Communication 3.4 Sever-Clients Architecture 3.5 Local Power Iteration with Warm Start 3.6 Weight Scaling Method 3.7 Fully Decentralized Architecture 3.8 Complexity Analysis 4 Experimental Results 4.1 Experiment on Structured Dataset 4.2 Case Studies 5 Concluding Remarks References Anchoring-and-Adjustment to Improve the Quality of Significant Features 1 Introduction 2 Proposed Two-Stage Anchoring-and-Adjustment Approach 2.1 Anchoring Stage Using Explainability Maximized Method 2.2 Adjustment Stage 3 Empirical Evaluation 4 Conclusion References Data Mining Based Artificial Intelligent Technique for Identifying Abnormalities from Brain Signal Data 1 Introduction 2 Workflow 2.1 Pre-processing the Brain Signal Data 2.2 Spectrogram Image Generation 2.3 Feature Extraction and Dimension Reduction 2.4 Classification of the Extracted Features 3 Performance Evaluation Materials and Parameters 3.1 Datasets 3.2 Classification Performance Measure 4 Experimental Results 4.1 Experimental Setup 4.2 Results 5 Conclusion References Where Should I Go? A Deep Learning Approach to Personalize Type-Based Facet Ranking for POI Suggestion 1 Introduction 2 Facet Ranking Related Research 3 Proposed Approach 4 Evaluation 5 Conclusions References Modeling Without Sharing Privacy: Federated Neural Machine Translation 1 Introduction 2 Proposed Method 2.1 Problem Definition 2.2 Architecture of FedNMT 2.3 Federated Vocabulary 2.4 Secure Model Training 2.5 Domain Expert 3 Experiment 3.1 Experiment Setups 3.2 Performance 4 Conclusion References Knowledge Graph and Entity Linking Encoding the Meaning Triangle (Object, Entity, and Concept) as the Semantic Foundation for Entity Alignment 1 Introduction 2 Problem Formulation 3 The C4EA Framework 3.1 Entity Representation Learning 3.2 Concept Representation Learning 3.3 Model Optimization 4 Experiments 4.1 Datasets 4.2 Experimental Setup 4.3 Experimental Results 4.4 Analysis 5 Related Work 6 Conclusion References Incorporating Network Structure with Node Information for Semi-supervised Anomaly Detection on Attributed Graphs 1 Introduction 2 Background 2.1 Problem Definition 2.2 Related Work 3 Proposed Method 3.1 Framework 3.2 Structure Anomaly Detection 3.3 Attribute Anomaly Detection 3.4 Loss Function 4 Experiment 4.1 Datasets 4.2 Baselines and Evaluation Metrics 4.3 Experimental Settings 4.4 Performance Evaluation 4.5 Parameter Sensitivity Analysis 5 Conclusion References OntoSP: Ontology-Based Semantic-Aware Partitioning on RDF Graphs 1 Introduction 2 Preliminaries 2.1 RDF and Graph Partitioning 2.2 Related Work 3 Ontology-Based Semantic-Aware Partitioning 3.1 Semantic Coherence 3.2 Adaptive Semantic Similarity 3.3 Initial Partitioning 3.4 The OntoSP Algorithm 4 Experiments 4.1 Datasets 4.2 Parameter Settings 4.3 Scalability 4.4 Efficiency 5 Conclusion 6 Appendix 6.1 Queries for DBpedia References Optimal Subgraph Matching Queries over Distributed Knowledge Graphs Based on Partial Evaluation 1 Introduction 2 Related Work 2.1 MapReduce-Based Graph Systems 2.2 Specialized RDF Systems 3 Preliminaries and Problem Statement 4 Partial Matching Index Based Algorithm 4.1 Partial Matching Index 4.2 PM-Index Based Assembly Algorithm 5 Experiments 5.1 Datasets and Queries 5.2 Experimental Results 6 Conclusion References Enhancing both Local and Global Entity Linking Models with Attention 1 Introduction 2 Related Work 2.1 Local Models 2.2 Global Models 3 Problem Definition 4 Model 4.1 Attention Based Local Model 4.2 Global Model with Document-Entity Coherence 5 Experiments 5.1 Implements 5.2 Datasets 5.3 Candidate Selection and Entity Embeddings 5.4 Hyper-parameter Setting 5.5 Results 5.6 Analysis 6 Conclusions and Future Work References HyperJOIE: Two-View Hyperbolic Knowledge Graph Embedding with Entities and Concepts Jointly 1 Introduction 2 Related Work 2.1 Single-View KGE Models 2.2 Two-View Joint KGE Models 2.3 Hyperbolic KGE Models 3 Problem Formulation and Background 3.1 Problem Formulation 3.2 Preliminaries on Hyperbolic Geometry 4 Methodology 4.1 Intra-view Model 4.2 Cross-View Model 4.3 Loss Function and Training 5 Experiments 5.1 Experimental Setup 5.2 KG Completion 5.3 Entity Typing 5.4 Case Study 6 Conclusion and Future Work References IOPE: Interactive Ontology Population and Enrichment Guided by Ontological Constraints 1 Introduction 2 Formal Background 2.1 RDF Format 2.2 Ontological Constraints 3 Interactive Ontology Update 3.1 The IOPEWeb Ontology 3.2 Ontology-Based GUI Construction 4 Evaluation 4.1 Evaluation of the IOPE GUI Efficiency 4.2 Evaluation of IOPE Users' Satisfaction 4.3 Effectiveness of IOPE for Enriching the OntoSAMSEI Ontology 5 Related Work 6 Conclusion References Graph Neural Network Controversy Detection: A Text and Graph Neural Network Based Approach 1 Introduction 2 Related Work 3 Graph Neural Network-Based Controversy Detection Approach 3.1 Graph Building 3.2 User Feature Extraction 3.3 Graph Embedding 3.4 Graph Classification 4 Experimental Evaluation 4.1 Dataset 4.2 Baseline 4.3 First Experiment: Controversy Detection Based on Structural Information 4.4 Second Experiment: Textual Content and Structural Information Based on Controversy Detection 5 Conclusion References GMGCN: Gated Memory Graph Convolutional Network for Passenger Demand Prediction 1 Introduction 2 Related Work 3 Methodology 3.1 Framework 3.2 Passenger Demand on Graph 3.3 Long-Term Temporal Attention 3.4 Gated Memory Spatial-Temporal Convolution (GMSTCN) Layer 3.5 Output Layer 4 Experiment 4.1 Datasets 4.2 Baselines 4.3 Experiment Settings 4.4 Results 4.5 Ablation Study 5 Conclusion References Event Detection in Social Media via Graph Neural Network 1 Introduction 2 Related Work 2.1 Event Detection 2.2 Short Text Classification 3 Methodology 3.1 Text-Level Graph Construction 3.2 Topic Extraction 3.3 Event Detection 4 Experiments 4.1 Datasets 4.2 Baseline Methods 4.3 Model Settings 4.4 Classification Results 4.5 Analysis of Topic Models 4.6 Analysis of the Number of Topics 5 Conclusion References Knowledge-Guided Fraud Detection Using Semi-supervised Graph Neural Network 1 Introduction 2 Methodology 2.1 Preliminaries 2.2 The Proposed Model: KS-GNN 3 Experiments 3.1 Experimental Setup 3.2 Performance Evaluation 3.3 Generalizations of Our Method 4 Conclusion References MSSF-GCN: Multi-scale Structural and Semantic Information Fusion Graph Convolutional Network for Controversy Detection 1 Introduction 2 Related Work 3 Methodology 3.1 TPCS Graph Construction 3.2 Multi-scale Structural and Semantic Information Fusion Graph Convolutional Network 4 Experiment 4.1 Datasets and Baselines 4.2 Performance Comparison 5 Conclusion References A Syntax-Aware Encoder for Authorship Attribution 1 Introduction 2 Related Work 3 Syntax-Aware Encoder Model 4 Experiments 5 Conclusion References Graph Attentive Leaping Connection Network for Chinese Short Text Semantic Classification 1 Introduction 2 Related Work 3 Graph Attentive Leaping Connection Model 3.1 Problem Definition 3.2 Model Description 4 Experiment 4.1 Experimental Setup 4.2 Evaluation Metrics and Baseline 4.3 Result and Analysis 5 Conclusion and Future Work References Graph Query Graph Ordering: Towards the Optimal by Learning 1 Introduction 2 Graph Ordering 3 Methodologies 3.1 Deep Order Network 3.2 RL-Sampler 4 Experimental Studies 5 Related Works 6 Conclusion References Fast Approximate All Pairwise CoSimRanks via Random Projection 1 Introduction 2 Preliminary 2.1 Notations and Terminology 2.2 Problem Definition 3 Related Work 4 The RPCS Algorithm 4.1 Main Algorithm 4.2 Analysis 4.3 Choosing 5 Experiments 5.1 Experimental Setting 5.2 Efficiency Evaluation 6 Conclusion References Critical Nodes Identification in Large Networks: An Inclination-Based Model 1 Introduction 2 Preliminaries 3 Solution 3.1 Baseline Algorithm 3.2 Follower Computation 3.3 Search Algorithm 4 Experiments 4.1 Experiment Setup 4.2 Effectiveness Evaluation 4.3 Efficiency Evaluation 5 Related Work 6 Conclusion References LPMA - An Efficient Data Structure for Dynamic Graph on GPUs 1 Introduction 2 Preliminary 2.1 Problem Definition 2.2 Existing CSR-Based Data Structure: GPMA+ 3 Leveled Packed Memory Array 3.1 Expansion in GPMA+ 3.2 LPMA Structure and Update 3.3 Analysis 4 Experimental Evaluation 4.1 Experimental Setup 4.2 Methodology 4.3 Expansion and Insertion Performance 4.4 Query Performance 5 Related Work References Updating Maximal (, )-Cliques of a Temporal Network Efficiently 1 Introduction 2 Preliminaries and Problem Definitions 3 Proposed Solution Approach 4 Experimental Evaluation References Social Network Web of Students: Class-Level Friendship Network Discovery from Educational Big Data 1 Introduction 2 Related Work 3 Problem Formulation 4 Design of CANDY 4.1 Feature Representation 4.2 Network Representation 4.3 Data Augmentation 4.4 Generative Model 4.5 Performance Evaluation Methodology 5 Dataset 6 Experiments 6.1 Experimental Settings 6.2 Analysis of Results 7 Conclusion References Event Cube for Suicidal Event Analysis: A Case Study 1 Introduction 2 Related Work 2.1 Event Detection 2.2 Event Relationship Analysis 2.3 Online Analytical Processing 3 The Event Cube Paradigm 3.1 Overview of the Framework 3.2 Data Collection 3.3 Data Consolidation 3.4 Data Fusion for Event Detection 3.5 Event Cube for Analysis 4 Case Study 4.1 Datasets 4.2 Main Architecture and Components 4.3 Case Study with EC Operations 5 Conclusion References Cross-modal Attention Network with Orthogonal Latent Memory for Rumor Detection 1 Introduction 2 Related Work 2.1 Multi-modal Rumor Detection 2.2 Multi-modal Data Fusion 3 Methodology 3.1 Overview of the Framework 3.2 Visual Extractor 3.3 Textual Extractor 3.4 Cross-modal Attention Fusion Network (CAF) 3.5 Rumor Detector 3.6 Loss Function 4 Experiments 4.1 Datasets 4.2 Baselines 4.3 Performance of the Approaches 4.4 Ablation Study 4.5 Effectiveness of CALM on Multimodal Fusion 4.6 Impact of the Number of Heads in CAF 4.7 Impact of the Number of Patterns in Latent Memory 4.8 Failure Cases Study 5 Conclusion References OMT: An Operate-Based Approach for Modelling Multi-topic Influence Diffusion in Online Social Networks 1 Introduction 2 Related Work 3 The Operator-Based Multiple-Topic (OMT) Modelling 3.1 Multi-Topic Influence Framework 3.2 Formal Definitions 3.3 Derivation 3.4 Network Diffusion Process 4 The Influence Maximization in Multiple-Topic Social Networks 5 Experiments and Discussion 5.1 Experiment Setup 5.2 Experiment 1 5.3 Experiment 2 5.4 Discussion 6 Conclusion References Modeling User Profiles Through Multiple Types of User Interaction Behaviors 1 Introduction 2 Overview of MMF 3 Application Example of MMF References HACK: A Hierarchical Model for Fake News Detection 1 Introduction 2 Related Work 2.1 Fake News Detection 2.2 Text Classification 3 Our Model 3.1 Character Feature Representation 3.2 Phrase Feature Representation 3.3 Sentence Feature Representation 4 Experiments 4.1 Datasets 4.2 Baselines 4.3 Experimental Results and Analysis 5 Conclusion References Spatial and Temporal Data Analysis An Efficient Approach for Spatial Trajectory Anonymization 1 Introduction 2 Related Work 3 Preliminaries 4 Methodology 4.1 Merge Loss 4.2 Gindex for Efficient Trajectory K-anonymization 5 Experiments 5.1 Results and Analysis 6 Conclusion References Developing a Deep Learning Based Approach for Anomalies Detection from EEG Data 1 Introduction 2 Proposed Framework 2.1 Collecting Raw EEG Data 2.2 Data Pre-processing (De-noising, Segmenting, and Down-Sampling) 2.3 Discover Hidden Significant Characteristics of Data and Classification Using GRU Based Scheme 2.4 Model's Performance Evaluation 3 Result and Analysis 4 Discussion 5 Conclusion References Dynamic Transit Flow Graph Prediction in Spatial-Temporal Network 1 Introduction 2 Related Work 3 Preliminaries 3.1 Transit Flow Graph 3.2 Node Flow 3.3 Neighborhood Relation Graph 3.4 Problem Statement 4 Methodology 4.1 Background Technologies 4.2 Spatial-Temporal Network 5 Experimental Settings 5.1 Datasets 5.2 Hyperparameters 5.3 Metrics 5.4 Compared Algorithms 6 Experimental Results 6.1 Performance Comparison 6.2 Effect of Input Sequence Length 6.3 Effect of Each Component 6.4 Effect of City Partition Methods 7 Conclusion References Disatra: A Real-Time Distributed Abstract Trajectory Clustering 1 Introduction 2 Related Works 3 Problem Definition 4 Real-Time Distributed Abstract Trajectory Clustering 4.1 Framework Overview 4.2 Trajectory Abstraction 4.3 GeoHash Indexing 4.4 Density-Based Line Clustering (DLC) 5 Experiment 5.1 Experiment Settings 5.2 Results and Analysis 6 Conclusion and Future Work References Extra-Budget Aware Task Assignment in Spatial Crowdsourcing 1 Introduction 2 Problem Definitions 3 Baseline Algorithms 3.1 The Optimal Solution 3.2 A Simple Greedy Algorithm 4 The Improved Greedy Algorithms 4.1 The Greedy Algorithm with Fewer Workers First 4.2 The Greedy Algorithm with Incremental Search 5 Experiments 5.1 Experimental Setup 5.2 Results on Real Dataset 6 Related Work 7 Conclusion References Expert Recommendations with Temporal Dynamics of User Interest in CQA 1 Introduction 2 Related Works 2.1 Expert Recommendation 2.2 Deep Neural Network 3 Our Approach 3.1 Question Encoder 3.2 User Encoder 3.3 Prediction and Training 4 Experiments 4.1 Dataset 4.2 Baseline 4.3 Evaluation and Implementation 4.4 Parameter Setting 4.5 Results Analysis 5 Conclusion References Author Index