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

Preface
Organization
Contents – Part III
Query Processing and Optimization
MSP: Learned Query Performance Prediction Using MetaInfo and Structure of Plans
	1 Introduction
	2 Related Work
	3 Feature Encoding
	4 Proposed Tree-structured Model
		4.1 Embedding Layer
		4.2 Representation Layer
		4.3 Prediction Layer
	5 Experiments
		5.1 Dataset
		5.2 Baselines
		5.3 Experiment Setting
		5.4 Experimental Results
		5.5 Ablation Analysis
		5.6 Model Adaptability
	6 Conclusion
	References
Computing Online Average Happiness Maximization Sets over Data Streams
	1 Introduction
	2 Related Work
	3 Problem Definition
	4 The GreedyAT Algorithm
		4.1 Properties
		4.2 The Algorithm
		4.3 Theoretical Analysis
	5 Experimental Evaluation
		5.1 Setup
		5.2 Experimental Results
	6 Conclusion
	References
A Penetration Path Planning Algorithm for UAV Based on Laguerre Diagram
	1 Introduction
	2 Problem Description
	3 A Penetration Path Planning Algorithm Based on Laguerre Diagram
		3.1 Route Chart Construction
		3.2 Path Planing
	4 Simulation and Analysis
	5 Conclusion
	References
Scheduling Strategy for Specialized Vehicles Based on Digital Twin in Airport
	1 Introduction
	2 Related Works
		2.1 Optimized Algorithms
		2.2 Off-Line Simulation
	3 Problem Statement
	4 Method
		4.1 Digital Twin Scheduling System Based on Edge-Computing
		4.2 Real-Time Scheduling Strategy Based-On Digital Twin
	5 Experiment
	6 Conclusion
	References
Recommender Systems
Graph-Based Sequential Interpolation Recommender for Cold-Start Users
	1 Introduction
	2 Preliminaries
		2.1 Problem Formulation
		2.2 Graph Models
	3 The Proposed Method
		3.1 Global Graph Layer
		3.2 Local Graph Layer
		3.3 Interpolation Layer
		3.4 Prediction Layer
	4 Experiments
		4.1 Datasets
		4.2 Baselines
		4.3 Experimental Setup
		4.4 Overall Performance (RQ1)
		4.5 Ablation Study
		4.6 Hyper-Parameter Study (RQ4)
	5 Related Work
		5.1 Sequential Recommendation
		5.2 Cold-Start Recommendation
	6 Conclusion
	References
Self-guided Contrastive Learning for Sequential Recommendation
	1 Introduction
	2 Related Work
		2.1 Sequential Recommendation
		2.2 Self-supervised Learning
	3 Method: Self-BERT
		3.1 Problem Definition
		3.2 Recommendation Task
		3.3 Self-guided Contrastive Learning Task
		3.4 Joint Learning
	4 Experiments
		4.1 Experimental Settings
		4.2 Overall Performance Comparison (RQ1)
		4.3 Parameter Sensitivity (RQ2)
		4.4 Ablation Study (RQ3)
		4.5 Robustness Analysis (RQ4)
	5 Conclusion
	References
Time Interval Aware Collaborative Sequential Recommendation with Self-supervised Learning
	1 Introduction
	2 Related Work
		2.1 Sequential Recommendation
		2.2 Self-supervised Learning in RS
	3 Problem Formulation
	4 Proposed Method
		4.1 Time Interval Aware Collaborative Model with Attention Adjustment
		4.2 Enhancing User Representation with Self-Supervised Learning
	5 Experiments
		5.1 Datesets
		5.2 Baselines and Evaluation Metrics
		5.3 Experimental Settings
		5.4 Experimental Results
		5.5 Ablation Study
		5.6 Case Study
		5.7 Hyperparameter Sensitivity
	6 Conclusion
	References
A2TN: Aesthetic-Based Adversarial Transfer Network for Cross-Domain Recommendation
	1 Introduction
	2 Related Work
		2.1 Cross-Domain Recommendation
		2.2 Adversarial Transfer Network
	3 Notations and Problem Definition
	4 The Proposed Model
		4.1 Aesthetic Feature Extraction Network
		4.2 General Feature Embedding Layer
		4.3 Adversarial Transfer Network
		4.4 Element-Wise Attention
		4.5 Model Learning
	5 Experiments
		5.1 Experimental Settings
		5.2 Performance Comparison
		5.3 Impacts of the Adversarial Transfer Network and Information Modeling
		5.4 Effects of the Hyper-Parameter
	6 Conclusion
	References
MORO: A Multi-behavior Graph Contrast Network for Recommendation
	1 Introduction
	2 Related Work
	3 Preliminary
	4 Methodology
		4.1 Behavior Propagation
		4.2 Behavior Perception
		4.3 Contrast Enhancement
		4.4 Multi-task Learning
	5 Experiments
		5.1 Experimental Setup
		5.2 Overall Performance (RQ1)
		5.3 Study of MORO (RQ2)
		5.4 Study of User Preferences (RQ3)
	6 Conclusion and Future Work
	References
Eir-Ripp: Enriching Item Representation for Recommendation with Knowledge Graph
	1 Introduction
	2 Related Work
	3 Problem Definition
	4 Model
		4.1 Overview
		4.2 User Interactive Learning Unit
		4.3 Enhancement Module of Item Representation
		4.4 RippleNet
		4.5 Forecast
		4.6 Loss Function
	5 Evaluation
		5.1 Datasets
		5.2 Baseline
		5.3 Experiment Setup
		5.4 Results
	6 Conclusion
	References
User Multi-behavior Enhanced POI Recommendation with Efficient and Informative Negative Sampling
	1 Introduction
	2 Related Work
		2.1 POI Recommendation
		2.2 Negative Sampling
	3 Preliminaries
	4 Methodology
		4.1 Context Graphs from User Multi-behavior
		4.2 Negative Sampling Based on Datum Points
		4.3 POI Recommendation with Multi-behavior
	5 Experiments
		5.1 Experimental Setting
		5.2 Performance Comparison
		5.3 Impacts of the Multi-behavior Data
		5.4 Impacts of Hyper-parameters
	6 Conclusion
	References
Neighborhood Constraints Based Bayesian Personalized Ranking for Explainable Recommendation
	1 Introduction
	2 Preliminary
		2.1 Explainability Matrix
		2.2 Explainable Bayesian Personalized Ranking
	3 Proposed Methods
		3.1 Constrained Explainable Bayesian Personalized Ranking (CEBPR)
		3.2 Adding Constraint for Explainable Negative Item (CEBPR+)
	4 Experiments
		4.1 Experiment Settings
		4.2 Results and Discussion
	5 Conclusion
	References
Hierarchical Aggregation Based Knowledge Graph Embedding for Multi-task Recommendation
	1 Introduction
	2 Related Work
	3 Our Approach
		3.1 User Feature Processing Unit
		3.2 Recommendation
		3.3 Hierarchical Aggregation Based Knowledge Graph Embedding
		3.4 Cross-compress Unit
		3.5 Optimization
	4 Experiments
		4.1 Datasets
		4.2 Experiments Setup
		4.3 Performance
	5 Conclusions
	References
Mixed-Order Heterogeneous Graph Pre-training for Cold-Start Recommendation
	1 Introduction
	2 Related Work
		2.1 Pre-training GNNs
		2.2 Cold-Start Recommendation
	3 The Proposed MHGP Model
		3.1 First-order Structure View Encoding
		3.2 High-order Structure View Encoding
		3.3 Pre-training with Contrastive Learning
		3.4 Fine-Tuning with Recommendation Models
	4 Experiments and Results
		4.1 Experiment Settings
		4.2 Overall Performance Comparison
		4.3 Ablation Study
	5 Conclusion and Future Work
	References
MISRec: Multi-Intention Sequential Recommendation
	1 Introduction
	2 Related Work
	3 Proposed Method
		3.1 Problem Setting
		3.2 Embedding Layer
		3.3 Multi-Intention Extraction Module
		3.4 Multi-Intention Evolution Module
		3.5 Multi-Intention Aggregation Module
		3.6 Model Training
	4 Experiments
		4.1 Experimental Setup
		4.2 Main Results
	5 Conclusion
	References
MARS: A Multi-task Ranking Model for Recommending Micro-videos
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Definition
		3.2 Model
	4 Experimental Evaluation
		4.1 Experimental Setup
		4.2 Performance Comparison
		4.3 Ablation Study
		4.4 Impact of Number of Experts
		4.5 Deployment and Online A/B Tests
	5 Conclusion
	References
Security, Privacy, and Trust and Blockchain Data Management and Applications
How to Share Medical Data Belonging to Multiple Owners in a Secure Manner
	1 Introduction
	2 Related Work
	3 Preliminaries
		3.1 MKE Scheme Based on the Chinese Remainder Theorem
		3.2 Logical Key Hierarchy
	4 System Model
		4.1 System Overview
		4.2 Transactions
		4.3 Signature and Verification
	5 Our Proposed Scheme
		5.1 Initialization
		5.2 Diagnosis
		5.3 Sharing
	6 Performance Analysis
		6.1 Security Features
		6.2 Master Key Encryption
		6.3 Blockchain-Based System
		6.4 Overhead of Sharing
	7 Conclusion
	References
LAP-BFT: Lightweight Asynchronous Provable Byzantine Fault-Tolerant Consensus Mechanism for UAV Network Trusted Systems
	1 Introduction
	2 System Model and Recommended Solution
		2.1 System Model
		2.2 Threat Model
		2.3 Blockchain Structure
		2.4 Lightweight Asynchronous Provable Byzantine Fault-Tolerant Consensus Mechanism
	3 Proof of System Properties
		3.1 Proof of Security
		3.2 Proof of Activity
	4 Simulation Experiments
		4.1 The Base Latency
		4.2 Latency Under Different Error Nodes
		4.3 The Rate of Energy Consumption
	5 Conclusion
	References
A Secure Order-Preserving Encryption Scheme Based on Encrypted Index
	1 Introduction
	2 Related Work
		2.1 Encrypted Database
		2.2 Operation on Encrypted Data
		2.3 Encrypted Index
	3 Secure Query Scheme
		3.1 Scheme Overview
		3.2 Order-Hiding Encrypted B+ Tree
		3.3 Query Algorithm Based on OHBPTree
	4 Security Analysis
		4.1 Security Proof
		4.2 Preventing Inference Attack
	5 Experiments
		5.1 Experiment Setup
		5.2 Evaluation
	6 Conclusion
	References
A Deep Reinforcement Learning-Based Approach for Android GUI Testing
	1 Introduction
	2 Background
	3 Approach
		3.1 Representation of States and Actions
		3.2 Reward Function
		3.3 Advantage Actor-Critic (A2C) Based Testing
	4 Empirical Study
		4.1 Applications Under Test
		4.2 Evaluation Setup
		4.3 Experiment Results
	5 Threats to Validity
	6 Related Work
	7 Conclusion
	References
RoFL: A Robust Federated Learning Scheme Against Malicious Attacks
	1 Introduction
	2 Related Works
	3 Problem Formulation
		3.1 Federated Learning Without Detection of Malicious Attacks
		3.2 Adversary Model
	4 Proposed Scheme
	5 Experiments and Analysis
		5.1 Experiments Settings
		5.2 Case 1, Experiments About RoFL with the Counter
		5.3 Case 2, Experiments About RoFL with the Validator
		5.4 Case 3, Experiments About RoFL with the Authenticator
	6 Conclusion
	References
FDP-LDA: Inherent Privacy Amplification of Collapsed Gibbs Sampling via Group Subsampling
	1 Introduction
	2 Preliminaries
	3 Framework of FDP-LDA
		3.1 Model Assumptions
		3.2 Guaranteed Privacy for Group Subsampling
		3.3 Guaranteed Privacy for FDP-LDA
	4 Experiments
		4.1 Inherent Privacy and Model\'s Utility
		4.2 Efficiency Improvements of FDP-LDA
	5 Related Work
	6 Conclusion and Future Work
	References
Multi-modal Fake News Detection Use Event-Categorizing Neural Networks
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Multi-modal Feature Extractor
		3.2 Event Categorizer
		3.3 Fake News Detector
		3.4 Model Integration
	4 Results
		4.1 Datasets
		4.2 Baselines
		4.3 Performance Comparison
	5 Conclusions
	References
Unified Proof of Work: Delegating and Solving Customized Computationally Bounded Problems in a Privacy-Preserving Way
	1 Introduction
	2 Preliminaries
		2.1 System Framework
		2.2 Security Model
		2.3 Cryptographic Primitives: Homomorphic Encryptions
	3 UPoW: All Put Together
	4 Experiments
		4.1 The Parameters for SEAL
		4.2 Computational Overhead
	5 Conclusion
	References
TSD3: A Novel Time-Series-Based Solution for DDoS Attack Detection
	1 Introduction
	2 DDoS Attack Detection Problem
	3 Novel Time-Series-Based Solution to DDoS Detection
		3.1 Motivation
		3.2 Overview of Proposed Approach
		3.3 Online Attention Sampling Algorithm
		3.4 Time-Series-Based Solution for DDoS Detection
		3.5 Analysis and Explanation
	4 Experiments and Results Analysis
		4.1 Experiment Setup
		4.2 Effects of System Parameters
		4.3 Performance Analysis
	5 Conclusion
	References
Executing Efficient Retrieval Over Blockchain Medical Data Based on Exponential Skip Bloom Filter
	1 Introduction
	2 Related Work
		2.1 Blockchain Applications for Medical Data Management Service
		2.2 Search Algorithm over Encrypted Data on Blockchain
	3 Problem Definition
		3.1 System Model
		3.2 Design Goals
	4 Preliminaries
		4.1 Bloom Filter
		4.2 Blockchain Technology
	5 Efficient Retrieval Algorithm for Blockchain Data
		5.1 Construction of Exponential Skip Bloom Filter Index Structure
		5.2 The Principle of the Efficient Query Algorithm over Blockchain
		5.3 Further Optimization of the Algorithm
		5.4 A Constant Value to Stop Endless Query
		5.5 Maintenance Mechanism of Index
	6 Performance and Security Analysis
		6.1 Performance Analysis
		6.2 Storage Overhead
		6.3 Secure Analysis
	7 Performance Evaluation
		7.1 Query Efficiency of Search Method
		7.2 Maintenance Overhead of Our Method
	8 Conclusion
	References
IMPGA: An Effective and Imperceptible Black-Box Attack Against Automatic Speech Recognition Systems
	1 Introduction
	2 Background and Related Work
		2.1 ASR Systems
		2.2 Adversarial Attacks on ASR Systems
	3 Imperceptible Genetic Algorithm (IMPGA) Attack Method
		3.1 Threat Model
		3.2 IMPGA Attack Method
		3.3 Gradient Estimation
	4 Evaluation
		4.1 Datasets and Evaluation Metrics
		4.2 Evaluation Results
		4.3 Ablation Study
	5 Conclusion
	References
FD-Leaks: Membership Inference Attacks Against Federated Distillation Learning
	1 Introduction
	2 Related Work
		2.1 Membership Inference
		2.2 Model Distillation
	3 Federated Distillation
	4 Client-Based Membership Inference
		4.1 Overview of the Attack
		4.2 Training the Attack Model
	5 Experimental Evaluation
		5.1 Datasets and Experimental Setup
		5.2 Impact of Model Distillation
		5.3 Effectiveness of Client-Based Membership Inference Attack
		5.4 Evaluation of the Number of Classes
	6 Summary and Future Work
	References
Spatial and Multi-media Data
When Self-attention and Topological Structure Make a Difference: Trajectory Modeling in Road Networks*-4pt
	1 Introduction
	2 Related Work
	3 Problem Definition
	4 Proposed Model TMRN
		4.1 Overview of TMRN
		4.2 Module Road2Vec
		4.3 Module LWA
		4.4 Module MOP
	5 Experiment
		5.1 Experimental Datasets
		5.2 Experimental Setup
		5.3 Evaluation Metric
		5.4 Compared Methods
		5.5 Impact of Different Modules
		5.6 Analysis of Parameters
	6 Conclusion and Future Work
	References
Block-Join: A Partition-Based Method for Processing Spatio-Temporal Joins
	1 Introduction
	2 Related Work
	3 Problem Definition
	4 The Framework
	5 Partitioned Blocks
		5.1 Equal-Sized Blocks
		5.2 Block Trajectories
		5.3 Unequal-Sized Blocks
		5.4 Adjacent Blocks
	6 Block-Join
		6.1 Filter and Refine Phase
		6.2 Algorithms
	7 Experimental Evaluation
		7.1 Setup
		7.2 Performance of Spatial-Temporal Join
	8 Conclusions
	References
RN-Cluster: A Novel Density-Based Clustering Approach for Road Network Partition
	1 Introduction
	2 Related Work
	3 Proposed Method
		3.1 Framework
		3.2 Density-Based Clustering of Road Graph
		3.3 The Attenuation of Neighborhood Connectivity (ANC)
	4 Experimental Results
		4.1 Evaluation Metrics
		4.2 Experimental Evaluation
		4.3 Visualization of Partition Results
	5 Conclusion
	References
Fine-Grained Urban Flow Inferring via Conditional Generative Adversarial Networks
	1 Introduction
	2 Related Work
		2.1 Image Super Resolution
		2.2 Urban Flow Analysis
	3 Formulation
	4 Methodology
		4.1 Adversarial Network Architecture
		4.2 External Factor Integration
		4.3 Flow Self-attention Module
	5 Experiment
		5.1 Dataset
		5.2 Experimental Settings
		5.3 Result on TaxiBJ
		5.4 Result on CCMobile
	6 Conclusion
	References
TDCT: Transport Destination Calibration Based on Waybill Trajectories of Trucks
	1 Introduction
	2 System Overview
	3 Demonstration
	References
OESCPM: An Online Extended Spatial Co-location Pattern Mining System
	1 Introduction
	2 System Overview
	3 Techniques
	4 Demonstration Scenarios
	5 Conclusion
	References
gTop: An Efficient SPARQL Query Engine
	1 Introduction
	2 System Structure
	3 Experiments and Demonstration
	References
Multi-SQL: An Automatic Multi-model Data Management System
	1 Introduction
	2 System Architecture and Implementation
	3 Key Technologies
	4 Demonstration Scenario
	References
Demonstration on Unblocking Checkpoint for Fault-Tolerance in Pregel-Like Systems
	1 Introduction
	2 Background
	3 Demonstration
	References
Author Index