دانلود کتاب Collaborative Computing: Networking, Applications and Worksharing: 18th EAI International Conference, CollaborateCom 2022 Hangzhou, China, October 15–16, 2022 Proceedings, Part I

فهرست مطالب :

Preface Conference Organization Contents – Part I Contents – Part II Recommendation System A Negative Sampling-Based Service Recommendation Method 1 Introduction 2 Related Work 2.1 GNN-Based Recommender System 2.2 Negative Sampling Method 3 Preliminaries 4 Method 4.1 Data Preprocessing 4.2 Synthetic Hard Negative Samples 4.3 Graph Neural Network 4.4 Service Recommendation 5 Experiments 5.1 Dataset Description 5.2 Evaluation Metrics 5.3 Baseline Methods 5.4 Experimental Performance 5.5 Hyperparameters Analysis 6 Conclusion and Future Work References Knowledge Graph Enhanced Web API Recommendation via Neighbor Information Propagation for Multi-service Application Development 1 Introduction 2 Related Works 2.1 Collaborative Filtering Based Web API Recommendation 2.2 Model Based Web API Recommendation 2.3 Deep Learning Based Web API Recommendation 2.4 Knowledge Graph Based Web API Recommendation 3 Preliminaries 3.1 Web API Knowledge Graph Construction 3.2 Problem Formulation 4 Methodology 4.1 Framework 4.2 Modeling 5 Experiments 5.1 Preparation 5.2 Comparison 5.3 Ablation Study 5.4 Impact of Hop Number and Neighbor Size 5.5 Impact of Embedding Size 5.6 Convergence of KGWARec 6 Conclusion References Expertise-Oriented Explainable Question Routing 1 Introduction 2 Related Works 2.1 Question Routing 2.2 Multi-task Learning 2.3 Explainable Recommendation 2.4 Compared with Existing Methods 3 Methods 3.1 Problem Definition 3.2 Question Title Encoder 3.3 Expertise-Oriented Expert Encoder 3.4 Multi-task Learning Framework 3.5 Differences with Existing Techniques 4 Experiments 4.1 Datasets and Experimental Settings 4.2 Performance Comparison 4.3 Ablation Study 4.4 Parameter Sensitivity Analysis 4.5 Case Study 5 Conclusion References An API Recommendation Method Based on Beneficial Interaction 1 Introduction 2 Related Work 2.1 Traditional Web Service Recommendation Method 2.2 Deep Learning-Based Web Service Recommendation 3 Method 3.1 Pre-processing 3.2 API Similarity Calculation 3.3 L0-sign Model Construction 4 Experiment 4.1 Dataset Description and Experimental Setup 4.2 Baselines 4.3 Expriment Results and Analysis 5 Conclusion and Future Work References A Flow Prediction Model of Bike-Sharing Based on Cycling Context 1 Introduction 2 Related Work 3 Analysis of Context Features 3.1 Analysis of Time Context 3.2 Analysis of Climate Context 4 Framework 4.1 Problem Definition 4.2 Feature Extraction of Time Series Based on LSTM 4.3 Computation of Important Features Based on Attention Mechanism 4.4 The Model Based on LSTM and Attention Mechanism 5 Experiments 5.1 Datasets 5.2 Setting 5.3 Evaluation Metrics 5.4 Results 6 Conclusion References Federated Learning and Application FedFR: Evaluation and Selection of Loss Functions for Federated Face Recognition 1 Introduction 2 Related Works 2.1 Face Recognition 2.2 Federated Learning 3 Classical Loss Functions 3.1 Classification-Based Loss Functions 3.2 Pair-Based Loss Functions 4 System Design and Performance Metrics 4.1 System Overview 4.2 Application Scenarios 4.3 Performance Metrics 5 Experimental Setup 5.1 Simulation Environment 5.2 Datasets and Preprocessing 5.3 Network Architectures 5.4 Implementation Details 6 Performance Evaluation and Summary 6.1 Comparison of Test Accuracy 6.2 Comparison of Communication Efficiency 6.3 Comparison of Local Training Time 6.4 Comparison of Memory Usage 6.5 Summary and Suggestion 7 Similarity Optimization Based Interpretation 8 Conclusion References FedCL: An Efficient Federated Unsupervised Learning for Model Sharing in IoT 1 Introduction 2 Related Work 3 Method 3.1 Federated Unsupervised Problem Definition 3.2 FedCL Overview 3.3 Federated Self-supervised Learning 3.4 Fine-Tuning and Distillation 4 Experiment 4.1 Implementation Details 4.2 Ablation Study 4.3 Performance Analysis 4.4 Comparison with Related Methods 4.5 Analysis of Scalability 5 Advantage and Limitation 6 Conclusion References Edge Federated Learning for Social Profit Optimality: A Cooperative Game Approach 1 Introduction 2 Preliminary 2.1 System Overview 2.2 Cooperative Games 2.3 Shapley Value 2.4 Shapley Value in EFL 3 Contribution and Aggregation Algorithms 3.1 Estimate of Shapley Value 3.2 IG-Shapley Value Based Federated Averaging (IG-Fedavg) 4 System Model 4.1 Income Model Based on IG-Shapley Value 4.2 System Latency of EFL 4.3 EFL Costs and Cooperation Benefits 5 Experiments 5.1 Simulation Setting 5.2 Performance Analysis of Shapley Value Based Training 5.3 Analysis of Social Benefit 6 Related Work 7 Conclusions References MetaEM: Meta Embedding Mapping for Federated Cross-domain Recommendation to Cold-Start Users 1 Introduction 2 Related Work 3 MODEL 3.1 Overview 3.2 Build MetaEM 3.3 Pretrain Stage 3.4 Mapping Stage 3.5 Cold-Start Stage 4 Experiments 4.1 Datasets 4.2 Implementation Details 4.3 Baselines 4.4 Evaluate Metrics 4.5 Experimental Results and Analysis 5 Conclusion References A Reliable Service Function Chain Orchestration Method Based on Federated Reinforcement Learning 1 Introduction 2 Related Work 3 Proposed Method 3.1 Service Function Chain Orchestration Model 3.2 Reinforcement Learning Model in Single Datacenter 3.3 Federated Reinforcement Learning Model in Multiple Datacenters 4 Simulation Analysis 4.1 Design of Simulation Experiment 4.2 Empirical Results 5 Conclusion References Edge Computing and Collaborative Working A Context-Aware Approach to Scheduling of Multi-Data-Source Tasks in Mobile Edge Computing 1 Introduction 2 Related Works 3 System Model and Problem Formulation 3.1 A Scenario of Multi-Data-Source Task 3.2 System Model 3.3 Task Model 3.4 Problem Formulation 4 Task Scheduling Scheme 4.1 Metadata Management 4.2 Data Management 4.3 Context-Aware Task Scheduling 4.4 Computation Offloading 5 Evaluation 5.1 Simulation Setup 5.2 Analysis of Data Storage and Management 5.3 Validation of Context-Aware Task Scheduling 5.4 Discussion on Computation Offloading 6 Conclusion References Secure and Private Coding for Edge Computing Against Cooperative Attack with Low Communication Cost and Computational Load 1 Introduction 2 Related Work 3 Problem Modeling 3.1 System Model 3.2 Attack Model, Secure Condition and Privacy Condition 3.3 Communication Cost and Computational Load 4 Secure and Private Coded Computation Schemes 4.1 Secure and Private Coding Scheme with Low Communication Cost (SPCC) 4.2 Secure and Private Coding Scheme with Low Computational Load (SPCL) 4.3 Example of SPCC and SPCL 5 Experiments 5.1 Parameter Settings 5.2 Experimental Results of Communication Cost 5.3 Experimental Results of Computational Load 5.4 Intuitive Understanding 6 Conclusion References Availability-Constrained Application Deployment in Hybrid Cloud-Edge Collaborative Environment 1 Introduction 2 Related Work 3 Problem Description 3.1 Resource Model 3.2 Time Model 3.3 Cost Model 3.4 Availability Constraints 3.5 Resource Constraints 3.6 Problem Definition 4 Algorithm Implementation 4.1 Chromosome Coding 4.2 Fitness Function and Constraints 4.3 Population Initialization 4.4 Selection Operator 4.5 Crossover Operator 4.6 Mutation Operator 4.7 Algorithm Description 5 Experimental Results and Analysis 5.1 Dataset 5.2 Simulation Settings and Parameter Settings 5.3 Results and Analysis 6 Conclusion References EBA: An Adaptive Large Neighborhood Search-Based Approach for Edge Bandwidth Allocation 1 Introduction 2 Related Work 3 System Model and Problem Formulation 3.1 System Model 3.2 Problem Formulation 4 Proposed EBA Approch 4.1 Feasible Initial Solution Generation 4.2 ALNS-Based Iterative Continuous Optimization 5 Experiments and Analysis 5.1 Experiment Settings 5.2 Performance Comparison 6 Conclusion and Future Work References System Completion Time Minimization with Edge Server Onboard Unmanned Vehicle 1 Introduction 2 Related Work 2.1 Static Edge Computation Offloading 2.2 Mobile Edge Computation Offloading 3 System Model 3.1 System Model 3.2 Computing Task Offloading Model 3.3 Communication Model 3.4 Computational Model 3.5 Mobile Model 4 Problem Formulation 4.1 Problem Overview 4.2 Constraint Analysis 4.3 Problem Formulation 5 Proposed Optimization Method 5.1 Community Partition 5.2 V-edge Location Optimization Method 6 Simulation Results 6.1 Evaluation Setup 6.2 Numerical Results 7 Conclusion References An Approach to the Synchronization of Dynamic Complex Network Combining Degree Distribution and Eigenvector Criteria 1 Introduction 2 Related Work 3 Overview Framework 3.1 Network Synchronization Problem 3.2 Our Method 4 Network Connection Based on Degree Distribution 4.1 Degree Distribution 4.2 The Method of DDC 4.3 The Algorithm of DDC 5 Enhanced Synchronization Small-World 5.1 Eigenvector Criterion 5.2 The Method of ESSW 5.3 The Algorithm of ESSW 6 Experiments 6.1 Evaluation Indexes 6.2 The Experiments of DDC 6.3 The Experiments of ESSW 6.4 Analysis of Influence of Network Topology on Synchronization Capability 7 Conclusions References An Energy-Saving Strategy for 5G Base Stations in Vehicular Edge Computing 1 Introduction 2 Related Works 3 System Model 3.1 Energy Model 3.2 Delay Model 4 Algorithms 4.1 Offline Strategy 4.2 Online Strategy 5 Simulation 5.1 Simulation Setup 5.2 Simulation Results 6 Conclusion References An Efficient Scheduling Strategy for Containers Based on Kubernetes 1 Introduction 2 Related Work 3 E-KCSS Scheduling Strategy 3.1 E-KCSS Architecture 3.2 Multicriteria Indicators 3.3 Adaptive Weight Mechanism 3.4 Load Balancing Strategy 4 Experimental Evaluation 4.1 Experimental Environment 4.2 Performance Indicators 4.3 Comparison of Cluster Resource Imbalance 4.4 Comparison of Deployment Efficiency 4.5 Comparison of Resource Utilization 5 Conclusion References s NOMA-Based Task Offloading and Allocation in Vehicular Edge Computing Networks 1 Introduction 2 Related Work 3 System Model and Problem Formulation 3.1 Transmission Model 3.2 Computation Model 3.3 Problem Formulation 4 Problem Analysis and Algorithm 4.1 Problem Analysis in Transmission Process 4.2 Problem Analysis in Computation Process 5 Simulation and Experiment 6 Conclusion References A Collaborative Graph Convolutional Networks and Learning Styles Model for Courses Recommendation 1 Introduction 2 Related Work 3 The Proposed Model 3.1 Problem Definition: Course Recommendation Models 3.2 Course Learning Styles Similarity Score 3.3 Graph Convolutional Network Course Prediction Rating 3.4 Collaborative Prediction Rating 4 Experiment Process 4.1 Experimental Environment and Data Set Processing 4.2 Evaluation Metrics 4.3 Comparison with Baseline Method (RQ1) 4.4 Experiments of Different Collaborative Weight Factor (RQ2) 4.5 Parameter Settings (RQ3) 5 Conclusion References Exploring the Impact of Structural Holes on the Value Creation in Service Ecosystems 1 Introduction 2 Value Creation Model of SE 3 Methodology 3.1 Research Questions 3.2 Impact Analysis Method 3.3 Concepts and Definitions 4 Experimental Design and Analysis 4.1 Construction of Computational Experiment 4.2 Experimental Design 4.3 Analysis on Individual Services Level (RQ1) 4.4 Analysis on SE Level (RQ2) 4.5 Answers and Analysis to the Research Questions 5 Related Work 6 Conclusion References Learning Dialogue Policy Efficiently Through Dyna Proximal Policy Optimization 1 Introduction 2 Preliminaries 2.1 Some Concepts of Pipeline System 2.2 Reinforcement Learning 3 DPPO Implementation 3.1 The Workflow 3.2 Direct Reinforcement Learning and Planning 3.3 World Model Training 3.4 World Model Deactivation 4 Experiments and Results 4.1 Dataset and User Simulator 4.2 Baselines 4.3 Results 5 Related Work 6 Conclusion References Self-gated FM: Revisiting the Weight of Feature Interactions for CTR Prediction 1 Introduction 2 Related Work 2.1 Factorization Machines 2.2 Attention Mechanism 2.3 Feature Selection for FMs 3 Methodology 3.1 Self-gated Factorization Machines 3.2 Self-gating Layer 3.3 Exploiting Self-gating Layer in Deep FM Models 3.4 Objective Function and Optimization 4 Experiment Settings 4.1 Datasets 4.2 Evaluation Methods 4.3 Evaluation Criteria 4.4 Implementation Details 5 Experiment Analysis 5.1 Migration to FM Models (RQ 1) 5.2 Performance Comparison (RQ 2) 5.3 Analysis of Influencing Factors (RQ 3) 5.4 Computational Efficiency (RQ 4) 6 Conclusions References Heterogeneous Graph Neural Network-Based Software Developer Recommendation 1 Introduction 2 Preliminary 3 Approach 3.1 Data Collection and Heterogeneous Graph Construction 3.2 Text Information Enhancement and Link Supplement 3.3 Meta-path and Self-supervised Joint Learning 3.4 Prediction Layer 3.5 Multi-task Training 4 Experiment Setup 4.1 Datasets and Metrics 4.2 Baselines and Hyper-parameters 4.3 Effectiveness of HGDR Compared to Other Baseline Methods (to Q1) 4.4 Ablation Studies (to Q2) 4.5 Effects of Different Meta-paths (to Q3) 4.6 Effects of Parameters (to Q4) 5 Related Work 5.1 Developer Recommendation 5.2 Heterogeneous Graph Neural Networks 6 Conclusion References Blockchain Applications FAV-BFT: An Efficient File Authenticity Verification Protocol for Blockchain-Based File-Sharing System 1 Introduction 2 Related Work 3 The Construction of FAV-BFT 3.1 Associating Identity and File Contents with VDF 3.2 Building a Challenge-Response Protocol Based on VDF 3.3 Verifying the File-Consistence 4 Evaluation of FAV-BFT 4.1 Performance Evaluation 4.2 Comparison with Multiple BFT Protocol 4.3 Security Analysis 5 Experiment and Evaluation Result 6 Conclusion References Incentive Mechanism Design for Uncertain Tasks in Mobile Crowd Sensing Systems Utilizing Smart Contract in Blockchain 1 Introduction 2 Related Work 3 Preliminaries 3.1 System Overview 3.2 Blockchain and Smart Contract 3.3 Design Objectives 4 Incentive Mechanism for Uncertain Scenario 4.1 Design Rationale 4.2 Design Details 4.3 Analysis 5 Performance Evaluation 5.1 Baseline Methods 5.2 Simulation Settings 5.3 Simulation Results 6 Conclusion References Research on the Update Method of CP-ABE Access Control Strategy Based on Smart Contract 1 Introduction 2 Related Work 3 Overview of Our Scheme 4 Implementation Details of Our Scheme 4.1 Comparison of CP-ABE Based on RSA and CP-ABE Based on Bilinear Mapping 4.2 Sample Smart Contract Design 5 Security and Performance Analysis of the Proposed Scheme 5.1 Security Analysis 5.2 Performance Analysis 6 Conclusion References Effective Blockchain-Based Asynchronous Federated Learning for Edge-Computing 1 Introduction 2 Related Work 3 Core Technology of FedLyra 3.1 Architecture 3.2 Workflow 3.3 Asynchronous Training and Aggregation 3.4 Reputation of Node 3.5 Council Selection and Consensus 4 Evaluation Results 4.1 Experiment Setting 4.2 Learning Performance of FedLyra 4.3 Impact of Reputation Threshold 5 Conclusion References One-Time Anonymous Certificateless Signcryption Scheme Based on Blockchain 1 Introduction 1.1 Related Work 1.2 Our Contribution 1.3 Organization 2 Preliminaries 2.1 Bilinear Mapping 2.2 Hard Problems 2.3 Smart Contract 2.4 Algorithm Model 2.5 Security Model 3 One-Time Anonymous Certificateless Signcryption Scheme 4 Security Analysis 4.1 Formal Security 4.2 Informal Security 5 Performance Analysis 5.1 Computation Cost 5.2 Communication Cost 5.3 Security Features Comparison 6 Conclusion References Author Index