Web and Big Data : 6th International Joint Conference, APWeb-WAIM 2022, Nanjing, China, November 25–27, 2022, Proceedings, Part I

This three-volume set, LNCS 13421, 13422 and 13423, constitutes the thoroughly refereed proceedings of the 6th International Joint Conference, APWeb-WAIM 2022, held in Nanjing, China, in August 2022. The 75 full papers presented together with 45 short papers, and 5 demonstration papers were carefully reviewed and selected from 297 submissions. The papers are organized around the following topics: Big Data Analytic and Management, Advanced database and web applications, Cloud Computing and Crowdsourcing, Data Mining, Graph Data and Social Networks, Information Extraction and Retrieval, Knowledge Graph, Machine Learning, Query processing and optimization, Recommender Systems, Security, privacy, and trust and Blockchain data management and applications, and Spatial and multi-media data. Preface Organization Contents – Part I Contents – Part II Contents – Part III Big Data Analytic and Management An Improved Yin-Yang-Pair Optimization Algorithm Based on Elite Strategy and Adaptive Mutation Method for Big Data Analytics 1 Introduction 2 Yin-Yang-Pair Optimization (YYPO) 2.1 Splitting Stage 2.2 Archive Stage 3 Improved Yin-Yang Pair Optimization (CM-YYPO) 3.1 Crossover Operator with Elite Strategy 3.2 Mutation Operator with Adaptive Mutation Probability 3.3 Splitting Method 3.4 Computational Complexity 4 Validation and Discussion 4.1 Experimental Results 4.2 Performance Analysis of CM-YYPO 5 Conclusions References MCSketch: An Accurate Sketch for Heavy Flow Detection and Heavy Flow Frequency Estimation 1 Introduction 2 Background 3 Design of MCSketch 3.1 Matthew Counter 3.2 Data Structure of MCSketch 4 Experiment Results 4.1 Experiment Setup 4.2 Experiment on Memory 4.3 Experiments on Threshold 5 Conclusion References The Influence of the Student's Online Learning Behaviors on the Learning Performance 1 Research Background 2 Related Research 2.1 Introduction to Learning Analytics 2.2 Research on the Relationship Between Online Learning Behavior and Learning Performance 2.3 Other Related Learning Analytics Methods 3 Platform Functions and Research Samples 4 Research Results 4.1 Data Preprocessing 4.2 Learning Analytics 5 Research Conclusions and Recommendations References A Context Model for Personal Data Streams 1 Introduction 2 The Situational Context 3 The Entity Type Graph 4 Case Study 5 Conclusion References ACF2: Accelerating Checkpoint-Free Failure Recovery for Distributed Graph Processing 1 Introduction 2 Background and Motivation 3 Partition-aware Backup Strategy 4 Incremental Protocol 5 System Implementation 6 Experiments 6.1 Experimental Setting 6.2 Efficiency of Partition-aware Backup Strategy 6.3 Efficiency of Incremental Protocol 7 Related Work 8 Conclusions References Death Comes But Why: An Interpretable Illness Severity Predictions in ICU 1 Introduction 2 Related Works 2.1 Severity Prediction on ICU Patients 2.2 Interpretable Deep Learning Models 3 Benchmark Dataset and Methods 3.1 MIMIC-III Dataset 3.2 Pre-processing 3.3 Prediction Models 3.4 Counterfactual Explanation 4 Experiment Result and Analysis 4.1 Model Performance 4.2 Explanation Cases 5 Conclusion and Future Work References Specific Emitter Identification Based on ACO-XGBoost Feature Selection 1 Introduction 2 Mathematical Model of Parameter Optimization and FS of XGBoost 3 SEI Based on ACO-XGBoost FS 3.1 Feature Extraction Based on Lifting Wavelet Package Transformation 3.2 Feature Importance Scoring Based on XGBoost 3.3 XGBoost Optimization Based on ACO 4 Experiment Procedure and Result Analysis 4.1 Datasets 4.2 Data Preprocessing 4.3 Feature Selection by Using Important Value of XGBoost 4.4 The Effect of Parameters on XGBoost 4.5 XGBOOSt’s Parameter Optimization Based on ACO 4.6 Performance Comparison 5 Conclusion References NED-GNN: Detecting and Dropping Noisy Edges in Graph Neural Networks 1 Introduction 2 Related Works 2.1 Graph Neural Networks 2.2 Graph Structure Modification 3 Case Study 4 Our Approach 4.1 Notations and Preliminaries 4.2 Noisy Edges Dropping 4.3 NED-GNN 4.4 The Variant of NED-GNN 5 Experiments 5.1 Experimental Settings 5.2 Experiments Comparison 5.3 Parameter Study 5.4 Training Visualization 6 Conclusion References A Temporal-Context-Aware Approach for Individual Human Mobility Inference Based on Sparse Trajectory Data 1 Introduction 2 Related Works 3 Preliminaries 4 Methodologies 5 Experiments 5.1 Datasets 5.2 Baselines 5.3 Experimental Setup 5.4 Empirical Analysis 6 Conclusion References Unsupervised Online Concept Drift Detection Based on Divergence and EWMA 1 Introduction 2 Related Work 2.1 Detection Algorithms Based on Error Rate 2.2 Detection Algorithms Using a Small Number of Labels 2.3 Detection Algorithms Using Divergence 3 The Proposed Method 3.1 Constructing Data Distribution Functions Based on Sliding Windows 3.2 Measuring Differences in Data Distribution Between Windows Using Jensen-Shannon Divergence 3.3 Calculating Concept Drift Threshold Using EWMA 4 Experiments 4.1 Datasets 4.2 Experimental Settings 4.3 Results and Analysis 5 Conclusion References A Knowledge-Enabled Customized Data Modeling Platform Towards Intelligent Police Applications 1 Introduction 2 Related Concepts and Running Example 2.1 Ontology 2.2 Query Expansion 2.3 Running Example 3 CPDMP Overview 3.1 Implementation of CPDMP 3.2 PSDKM 3.3 CPDMP 3.4 Methodology 4 Experiments 4.1 Experiments Setting 4.2 Data Set 4.3 Experiment Results and Analysis 5 Related Work 6 Conclusion References Integrated Bigdata Analysis Model for Industrial Anomaly Detection via Temporal Convolutional Network and Attention Mechanism 1 Introduction 2 Related Works 2.1 Anomaly Detection 2.2 Application of Temporal Convolutional Network 3 IAD-TCN Framework 3.1 Temporal Convolutional Network to Capture Time Series Dependence 3.2 Attention Mechanism for Improved Model 4 Experimental Results and Analysis 4.1 Datasets Description 4.2 Evaluation Metrics 4.3 Anomaly Detection Results 5 Conclusions References Advanced Database and Web Applications WSNet: A Wrapper-Based Stacking Network for Multi-scenes Classification of DApps 1 Introduction 2 Related Work 3 Preliminaries 3.1 DApps Background 3.2 Problem Definition 3.3 Limitation of Existing Methods 4 WSNet 4.1 Feature Extraction 4.2 Ensemble Learning 5 Experiments 5.1 Dataset 5.2 Experiments Settings 5.3 Experimental Results 5.4 Ablation Studies 6 Conclusion References HaCache: A Hybrid Adaptive Cache for Persistent Memory Based Key-Value Systems 1 Introduction 2 Background and Motivation 2.1 PM-Based KV Systems 2.2 KV, KP and Block Cache 3 Related Work 4 Design and Implementation 4.1 Basic Operations of HaCache 4.2 Adaptive Adjustment Within Point Cache 4.3 Adaptive Adjustment Between Point Cache and Block Cache 5 Evaluation 5.1 Impact of Cache Size 5.2 Adaptive Adjustment 6 Conclusion References An Energy-efficient Routing Protocol Based on Two-Layer Clustering in WSNs 1 Introduction 2 Related Work 3 Description of DKBDCERP 3.1 First-level Cluster 3.2 Second-level Cluster 3.3 Inter-cluster Routing 4 Simulation and Performance Analysis 5 Conclusion References An Energy-efficient Routing Protocol Based on DPC-MND Clustering in WSNs 1 Introduction 2 Related Work 3 The Description of DMBERP Protocol 3.1 Cluster Establishment Phase 3.2 Cluster Head Election Phase 3.3 Inter-cluster Routing Establishment Phase 4 Simulation and Performance Analysis 5 Conclusion References Cloud Computing and Crowdsourcing Lightweight Model Inference on Resource-Constrained Computing Nodes in Intelligent Surveillance Systems 1 Introduction 2 Related Work 2.1 Intelligent Surveillance Systems 2.2 Multiple Targets Multiple Cameras Tracking (MTMCT) 2.3 Collaborative Learning Based ISSs 3 System Model and Problem Formulation 3.1 System Model 3.2 Horizontal Partition Algorithm 3.3 Vertical Partition Algorithm 4 Algorithm Design 5 Numerical Evaluation 5.1 Experiment Setup 5.2 Experimental Result 6 Conclusion References Trajectory Optimization for Propulsion Energy Minimization of UAV Data Collection 1 Introduction 2 Problem Formulation and System Model 2.1 Reliable Communication Distance 2.2 Propulsion Energy Minimization Model 3 Propulsion Energy Minimization Scheme 3.1 Virtual Base Station Deployment 3.2 Trajectory Initialization and Optimization 3.3 Polling Mechanism 4 Numerical Results 5 Conclusion References Robust Clustered Federated Learning with Bootstrap Median-of-Means 1 Introduction 2 Related Work 2.1 Clustered Federated Learning 2.2 Model Poisoning Attacks in Federated Learning 2.3 Robust Clustering 3 Methodology 3.1 Problem Definition 3.2 Robust Clustered FL with bMOM 3.3 Algorithm 4 Experiment 4.1 Training Setups 4.2 Experiment Analysis 5 Conclusion and Remarks References SAPMS: A Semantic-Aware Privacy-Preserving Multi-keyword Search Scheme in Cloud 1 Introduction 2 Notations and Preliminaries 2.1 Notations 2.2 Preliminaries 3 Problem Statement 4 The Proposed Search Scheme 4.1 Data Preprocessing Module 4.2 Index Building Module 4.3 Multi-keyword Search Module 5 Performance Evaluation 5.1 Semantic Precision Evaluation 5.2 Search Time Cost Evaluation 6 Conclusion References Task Assignment with Spatio-temporal Recommendation in Spatial Crowdsourcing 1 Introduction 2 Problem Definition 3 Methodology 3.1 Framework Overview 3.2 Spatio-temporal Preference Modeling 3.3 Task Assignment 4 Experimental Evaluation 4.1 Experimental Setup 4.2 Experimental Results 5 Related Work 6 Conclusion References DE-DQN: A Dual-Embedding Based Deep Q-Network for Task Assignment Problem in Spatial Crowdsourcing 1 Introduction 2 Problem Statement and Preliminaries 3 DE-DQN: Dual-Embedding Based Deep Q-Network for Task Assignment 3.1 Dual-Embedding 3.2 Framework and Training for DE-DQN 4 Experiments 4.1 Experiment Setups 4.2 Performance Analysis 5 Related Work 6 Conclusion References Dynamic Vehicle-Cargo Matching Based on Adaptive Time Windows 1 Introduction 2 Related Work 3 Problem Statement 4 Methodology 4.1 Loading Plan Generating 4.2 Adjusting Strategy Learning of Time Window Size 4.3 Matching Decision Making 5 Experiments 5.1 Data Description 5.2 Evaluation Metric 5.3 Comparative Approaches 5.4 Quantitative Comparison Results 5.5 Case Study 6 Conclusions References Microservice Workflow Modeling for Affinity Scheduling to Improve the QoS 1 Introduction 2 Background and Related Work 2.1 Affinity Scheduling 2.2 Microservice Workflow 3 Microservice Workflow Modeling 3.1 Definition of the Microservice Workflow 3.2 Running Time Estimation 3.3 Affinity Degree Calculation 4 Experimental Evaluation 4.1 Experiment Setting 4.2 Accuracy Validation on the Model 4.3 Effectiveness Validation on Performance Optimization Method 5 Conclusion References Data Mining SynBERT: Chinese Synonym Discovery on Privacy-Constrain Medical Terms with Pre-trained BERT 1 Introduction 2 Concept Introduction 3 Framework Introduction 3.1 Training Binary Classifier 3.2 Synonym Sets Discovery 4 Experiment 4.1 Baseline Methods 4.2 Dataset Construction 4.3 Terms Embedding and Parameter Setting 4.4 Evaluation Metrics 4.5 Experiment Results 5 Related Work 5.1 Medical Entity Linking 5.2 Entity Synonym Discovery 6 Conclusion References Improving Motor Imagery Intention Recognition via Local Relation Networks 1 Introduction 2 Related Work 3 MI Intention Recognition via LR-Net 3.1 Data Acquisition and Preprocessing 3.2 Generating Images from EEG Signals 3.3 Architecture 4 Experiments 4.1 Datasets and Settings 4.2 Benchmark Methods 4.3 Results and Discussion 5 Conclusion References EAS-GCN: Enhanced Attribute-Aware and Structure-Constrained Graph Convolutional Network 1 Introduction 2 Related Work 2.1 Graph Neural Networks 2.2 Auto-Encoder 2.3 Self-supervised Learning 3 The Proposed Model 3.1 Preliminaries 3.2 Auto-Encoder Module 3.3 Degree Prediction Module 3.4 Delivery Mechanism 3.5 Theory Analysis 4 Experiments 4.1 Datasets 4.2 Implementation and HyperParamter 4.3 Classification Results 4.4 Training Time Comparison 4.5 Ablation Study 4.6 Parameter Sensitivity Analysis 4.7 Over-Smoothing Analysis 4.8 Visualization 5 Conclusions References Mining Frequent Patterns with Counting Quantifiers 1 Introduction 2 Preliminaries 2.1 Conventional Graph Pattern Mining 2.2 Quantified Graph Patterns 3 Frequent QGPs Mining 3.1 Identifying Frequent QGPs 4 Experimental Study 5 Conclusion References MST-GNN: A Multi-scale Temporal-Enhanced Graph Neural Network for Anomaly Detection in Multivariate Time Series 1 Introduction 2 Related Work 2.1 Traditional Anomaly Detection 2.2 Deep Learning-Based Anomaly Detection 3 Methodology 3.1 Task Definition and Notations 3.2 Model Architecture 3.3 Multi-scale Temporal-Enhanced Representation 3.4 MST Guided Graph Construction 3.5 Graph Attention-Based Learning 3.6 Forecasting and Anomaly Detection 4 Experiments 4.1 Datasets and Evaluation Metrics 4.2 Performance Comparison 4.3 Discussion 5 Conclusion References Category Constraint Spatial Keyword Preference Query Based Spatial Pattern Matching 1 Introduction 2 Related Works 3 Problem Definition 4 CSF Algorithm 5 Experiments 6 Conclusion References Many-to-Many Pair Trading 1 Introduction 2 Related Work 3 Many-to-Many Pair Formation for Pair Trading 3.1 Problem Description 3.2 Pair Formation Framework 4 Experiments and Discussions 4.1 Datasets and Experimental Settings 4.2 Results and Analysis 5 Conclusion and Future Work References Proximity Preserving Graph Convolutional Networks 1 Introduction 2 Problem Formalization 3 Proximity Preserving Graph Convolutional Networks 3.1 k-order Smoothness of Graphs 3.2 Proximity Preserving Graph Convolutional Networks 3.3 Connections to Squared-Error P-reg 4 Experiments 4.1 Datasets and Experimental Settings 4.2 Node Classification Accuracy of PPGCNs 5 Conclusion References Discovering Prevalent Weighted Co-Location Patterns on Spatial Data Without Candidates 1 Introduction 2 Related Work 3 Method 3.1 Related Definitions 3.2 Querying Mining Framework Without Candidates 4 Evaluation 4.1 The Number of Mined Patterns 4.2 The Efficiency of the Querying Mining Framework 5 Conclusion References A Deep Looking at the Code Changes in OpenHarmony 1 Introduction 2 Methodology 2.1 Dataset 2.2 Research Questions 2.3 Descriptive Statistics on the Data 3 Empirical Results 3.1 [RQ-1:] Fault Distribution 3.2 [RQ-2:] File Type for Modifications 3.3 [RQ-3:] Additions and Deletions 4 Related Work 5 Conclusions References GADAL: An Active Learning Framework for Graph Anomaly Detection 1 Introduction 2 Related Works 3 Problem Formulation 4 Proposed Method 4.1 Abnormal-aware Query Strategy 4.2 Inconsistency-Aware Query Strategy 4.3 Integrating Query Strategies 5 Experiments 6 Conclusion References Fake News Detection Based on the Correlation Extension of Multimodal Information 1 Introduction 2 Related Work 2.1 Image Analysis and Text Classification 2.2 Fake News Detection 3 Our Approach 3.1 Problem Definition 3.2 Visual Feature Extractor 3.3 News Text Feature Extractor 3.4 OCR Text Feature Extractor 3.5 Text Correlation Extractor 3.6 Feature Integration Classifier 4 Experiments 4.1 Dataset 4.2 Baseline 4.3 Parameter Setting 4.4 Comparison Results 5 Conclusion References SPSTN: Sequential Precoding Spatial-Temporal Networks for Railway Delay Prediction 1 Introduction 2 Related Work 3 Methodology 3.1 Problem Formulation 3.2 Our Proposed Model 4 Experiments 4.1 Evaluation Data and Metrics 4.2 Results and Analysis 5 Conclusion References Graph Data and Social Networks Mining Periodic k-Clique from Real-World Sparse Temporal Networks 1 Introduction 2 Preliminaries 3 The Proposed Algorithms 3.1 Augmented Adjacency Array 3.2 The Proposed Pruning Techniques 3.3 Enumerating Periodic k-Cliques 4 Experiments 5 Related Work 6 Conclusion References LSM-Subgraph: Log-Structured Merge-Subgraph for Temporal Graph Processing 1 Introduction 2 Background and Related Work 2.1 Graph Storage Formats 2.2 Packed Memory Arrays 2.3 Temporal Graph Storage 3 LSM-Subgraph Design 3.1 Overview 3.2 PMA-Based Adjacency Array Model 3.3 Fluctuation-Aware Snapshot Creation Method 3.4 Log-Merging Method 4 Evaluation 4.1 Experiment Setup and Datasets 4.2 Overall Performance Comparison 4.3 Comparison Within LSM-Subgraph 4.4 System Design Parameters 5 Conclusion References ForGen: Autoregressive Generation of Sparse Graphs with Preferential Forest 1 Introduction 2 Related Work 3 Methodology 3.1 Preliminaries 3.2 Overview 3.3 A Hierarchical Tree Structure for Edge Generation 3.4 Two-Level Autoregressive Model 4 Experiments 4.1 Datasets 4.2 Settings 4.3 Model Effectiveness 4.4 Model Efficiency 4.5 Ablation Study 5 Conclusion References Iterative Deep Graph Learning with Local Feature Augmentation for Network Alignment 1 Introduction 2 Related Work 2.1 Spectral Method 2.2 Network Representation Learning Method 3 Preliminaries 3.1 Problem Formulation 3.2 Knowledge Representation Model 4 Method 4.1 IDGL-Based Node Embedding 4.2 Model Training 4.3 Alignment Prediction 5 Experiment 5.1 Experiment Setup 5.2 Experiment Result 5.3 Hyperparameter Sensitivity 6 Conclusion References OntoCA: Ontology-Aware Caching for Distributed Subgraph Matching 1 Introduction 2 Related Work 3 Preliminaries 4 Ontology-Aware Caching 4.1 Caching Schema 4.2 Workload-Adaptive Prefetching Strategy 4.3 Query Processing with OntoCA 5 Experiments 5.1 Experimental Settings 5.2 Experimental Results 6 Conclusion References A Social-Aware Deep Learning Approach for Hate-Speech Detection 1 Introduction 2 Methodology 2.1 CNN Architecture 2.2 LSTM Architecture 2.3 Social-Aware Deep Learning Network 3 Dataset 4 Experiments and Results 5 Conclusion References Community Detection Based on Deep Dual Graph Autoencoder 1 Introduction 2 Related Work 3 Preliminary 4 Proposed Method 4.1 Deep Dual Graph Autoencoder 4.2 Clustering Layer 4.3 Joint Optimization 5 Experiment 5.1 Experimental Settings 5.2 Experiment Result 5.3 Ablation Study 6 Conclusion and Further Work References SgIndex: An Index Structure Supporting Multiple Graph Queries 1 Introduction 2 Related Work 3 Index Based on Subgraph 3.1 Index Structure and Establishment Method 3.2 Path Query Algorithm 3.3 Subgraph Matching Algorithm 4 Experiments 4.1 Experimental Environment and Dataset 4.2 Path Query 4.3 Subgraph Matching 5 Conclusion References Author Index