Collaborative Computing: Networking, Applications and Worksharing: 17th EAI International Conference, CollaborateCom 2021, Virtual Event, October 16-18, 2021, Proceedings, Part I

Preface
Organization
Contents – Part I
Contents – Part II
Optimization for Collaborate System (Workshop Papers)
Chinese Named Entity Recognition Based on Dynamically Adjusting Feature Weights
	1 Introduction
	2 Model Approach
		2.1 BERT
		2.2 CNN
		2.3 BILSTM
		2.4 CRF
	3 BERT+EL-LGWF+CRF
		3.1 Weighted Fusion According to CNN and BILSTM
		3.2 EL
		3.3 Loss Function
	4 Experiment and Analysis
		4.1 Dataset
		4.2 Evaluation Indices
		4.3 Experimental Results and Analysis
	5 Summary
	References
Location Differential Privacy Protection in Task Allocation for Mobile Crowdsensing Over Road Networks
	Abstract
	1 Introduction
	2 Preliminary
		2.1 Privacy Model
		2.2 Adversary Model
	3 PPTA Framework
		3.1 Location Obfuscation
		3.2 Task Allocation Based on Obfuscated Locations
		3.3 Speed-Up with δ-Spanner Graph
	4 Evaluation
		4.1 Experiment Configurations
		4.2 Experimental Results
	5 Related Work
	6 Conclusion
	Acknowledgments
	References
“Failure” Service Pattern Mining for Exploratory Service Composition
	Abstract
	1 Introduction
	2 Related Work
		2.1 Log-Based Service Pattern Mining
		2.2 Process-Based Service Pattern Mining
	3 Model Definition
		3.1 Exploratory Service Composition Instance Model
		3.2 Service Pattern Model
	4 FSPMA
	5 Prototype Implementation
	6 Experiment
		6.1 Dataset and Environment
		6.2 Experimental Verification
	7 Application
		7.1 Service Recommendation Using “Failure” Service Patterns
		7.2 An Example
	8 Conclusion
	Acknowledgements
	References
Optimal Control and Reinforcement Learning for Robot: A Survey
	1 Introduction
	2 Optimal Control Problem Statement
	3 Solutions of Optimal Control for Robot
		3.1 Overview the Related Approaches
		3.2 Improve Precision and System Complexity
		3.3 Overcome Model Bias
		3.4 Reduce Computation
	4 Future Prospects and Discussion
	5 Summary and Conclusions
	References
KTOBS: An Approach of Bayesian Network Learning Based on K-tree Optimizing Ordering-Based Search
	Abstract
	1 Introduction
	2 Bayesian Network and k-tree
		2.1 Bayesian Network
		2.2 Tree Width and k-tree
	3 BN Learning Based on K-tree Optimizing Ordering-Based Search
		3.1 Obtaining Candidate Parent Set
		3.2 Initial Network Construction
		3.3 Optimizing Network Using Ordering-Based Search
	4 Experiment
		4.1 Dataset and Evaluation Method
		4.2 Experiment Results and Analysis
	5 Conclusions
	Acknowledgement
	References
Recommendation Model Based on Social Homogeneity Factor and Social Influence Factor
	Abstract
	1 Introduction
	2 Related Work
	3 Preliminary and Problem Definition
		3.1 Novel Graph Attention Network
		3.2 Notations
		3.3 Problem Definition
	4 The Proposed Model
		4.1 Model Details
			4.1.1 Original Input and Similar Item Network
			4.1.2 Embedding Layer
			4.1.3 NGAT Layer
			4.1.4 Pairwise Neural Interaction Layer
			4.1.5 Policy-Based Fusion Layer
			4.1.6 Output Layer and Loss Function
		4.2 Model Training
			4.2.1 Mini-Batch Training
			4.2.2 Alleviate Overfitting
	5 Experiments
		5.1 Dataset Introduction
		5.2 Experimental Setup
			5.2.1 Experimental Environment Setting
			5.2.2 Evaluation Metrics
			5.2.3 Compare Models
		5.3 Comparative Experiments: RQ1
		5.4 Ablation Experiments: RQ2
		5.5 Parameter Sensitivity Experiments: RQ3
	6 Conclusion
	References
Attention Based Spatial-Temporal Graph Convolutional Networks for RSU Communication Load Forecasting
	1 Introduction
	2 Related Works
	3 System Model
	4 Communication Load Evaluation
	5 The Communication Load Prediction Model
		5.1 Temporal Attention
		5.2 Spatial Attention
		5.3 Graph Convolution in Spatial Dimension
		5.4 Convolution in Temporal Dimension
		5.5 Fully Connected Layer
	6 Simulation and Analysis
		6.1 Dataset
		6.2 Model Parameters
		6.3 Comparison and Result Analysis
	7 Conclusion
	References
UVA and Traffic System
Mobile Encrypted Traffic Classification Based on Message Type Inference
	1 Introduction
	2 Related Work
		2.1 Traditional Unencrypted Traffic Classification
		2.2 Sequence Feature-Based Encrypted Traffic Classification
		2.3 Attribute Feature-Based Encrypted Traffic Classification
	3 System Introduction
		3.1 System Overview
		3.2 Data Preprocessing
		3.3 Message Type Inference
		3.4 Machine Learning
	4 Evaluation
		4.1 Preliminary
		4.2 Analysis of the Message Type Inference
		4.3 Analysis of Adopted Features
		4.4 Comparisons with Existing Approaches
	5 Discussion and Conclusion
	References
Fine-Grained Spatial-Temporal Representation Learning with Missing Data Completion for Traffic Flow Prediction
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Feature Extractors
		3.2 Data Completer
	4 Experiments and Analysis
		4.1 Experimental Settings
		4.2 Performance of Traffic Flow Prediction
		4.3 Effect of Data Completer
	5 Conclusion and Future Work
	References
Underwater Information Sensing Method Based on Improved Dual-Coupled Duffing Oscillator Under Lévy Noise Description
	Abstract
	1 Introduction
	2 Related Work
		2.1 Lévy Noise Model
		2.2 Chaotic Oscillator Signal Sensing System
	3 Approach
		3.1 Lévy Noise Model Describes Underwater Natural Environment Interference
		3.2 Improved Signal Sensing Method of Dual Coupling Duffing Oscillator
	4 Experiment and Analysis
		4.1 Experiment Deployment
		4.2 Performance
	5 Conclusion
	Acknowledgement
	References
Unpaired Learning of Roadway-Level Traffic Paths from Trajectories
	1 Introduction
	2 Related Work
	3 Preliminary Concepts
	4 Overview
		4.1 Definition
		4.2 Problem Analysis and Approach Overview
	5 Trajectory Data Transition
		5.1 Feature Extraction
		5.2 Orientation Converted to Color Information
	6 Training Model
	7 Experiment and Analysis
		7.1 Dataset and Experimental Environment
		7.2 Parameter Setting and Data Division
		7.3 Results and Performance Comparison
	8 Conclusion
	References
Multi-UAV Cooperative Exploring for the Unknown Indoor Environment Based on Dynamic Target Tracking
	1 Introduction
	2 Related Work and Scenario Description
		2.1 Related Work
		2.2 Scenario Description
	3 Method
		3.1 Wall-Around Algorithm
		3.2 Tracking-D*Lite
	4 Experiments
		4.1 Tracking-D*Lite Algorithm Experiment
		4.2 Simulation Experiment
	5 Conclusion
	References
Recommendation System
MR-FI: Mobile Application Recommendation Based on Feature Importance and Bilinear Feature Interaction
	Abstract
	1 Introduction
	2 Related Work
	3 Proposed Method
		3.1 Embedding Layer
		3.2 SENET Layer
		3.3 Bilinear-Interaction Layer
		3.4 Connectivity Layer
		3.5 Deep Network
		3.6 Prediction Layer
	4 Experimental Result and Analysis
		4.1 Data Set and Experiment Setup
		4.2 Evaluation Metrics
		4.3 Baseline Methods
		4.4 Experimental Performance
		4.5 Hyperparameters Analysis
	5 Conclusion and Future Work
	Acknowledgement
	References
Dual-Channel Graph Contextual Self-Attention Network for Session-Based Recommendation
	Abstract
	1 Introduction
	2 Related Work
		2.1 Neural Network Model
		2.2 Attention Mechanism
	3 Proposed Method
		3.1 Problem Statement
		3.2 Model Overview
		3.3 Session Graph Construction
		3.4 Item Embedding Learning
		3.5 Self-attention Network
		3.6 Prediction Layer
	4 Experiments and Analyses
		4.1 Datasets
		4.2 Evaluation Metric
		4.3 Experiment Settings
		4.4 Comparison with Baseline Methods
		4.5 The Influence of Model Parameters on Experimental Results
	5 Conclusions
	References
Context-aware Graph Collaborative Recommendation Without Feature Entanglement
	1 Introduction
	2 Problem Formulation
	3 Methodology
		3.1 General Framework
		3.2 Optimization
		3.3 Time Complexity Analysis of CGCR
	4 Experiments
		4.1 Dataset Description
		4.2 Experimental Settings
		4.3 RQ1: Does the Proposed Method Perform Better Than Other Comparison Methods?
		4.4 RQ2: Does the Proposed Method Elucidate the Meanings of Each Dimension of the Embedding?
		4.5 RQ3: How Does Number of Graph Convolution Layer Impact?
		4.6 RQ4: How Does Non-sampling Strategy Impact?
	5 Related Work
		5.1 Collaborative Filtering
		5.2 Non-sampling Learning for Top-K Recommendation
	6 Conclusion and Future Work
	References
Improving Recommender System via Personalized Reconstruction of Reviews
	1 Introduction
	2 Related Work
		2.1 Review-based Recommendation with Topic Modeling
		2.2 Document-Level Recommendation
		2.3 Review-Level Recommendation
	3 Methodology
		3.1 Probem Definition
		3.2 Overall Framework of PPRR
		3.3 Review Document Reconstruction Network (Re-Doc-Net)
		3.4 Document-Level Encode Network(Doc-Net)
		3.5 Review-Level Encode Network(Review-Net)
		3.6 Rating Score Prediction Layer
	4 Experiments and Analysis
		4.1 DataSets and Experiments Settings
		4.2 Performance Evaluation
		4.3 Discussion
		4.4 Hyper-Parameters Analyses
	5 Conclusion
	References
Recommendation System and Network and Security
Dynamic Traffic Network Based Multi-Modal Travel Mode Fusion Recommendation
	Abstract
	1 Introduction
	2 Concepts Used in the Paper
		2.1 Definition of the Fusion Recommendation Problem
		2.2 Heterogeneous Transport Travel Networks
		2.3 Meta-paths Extraction Based on User Trajectory
		2.4 Meta-path Guided Neighbors
	3 Heterogeneous Transport Travel Network Recommendation Model
		3.1 Initial Embedding
		3.2 Practice of Meta-path
		3.3 Meta-path Aggregation Functions
		3.4 Semantic Aggregation
		3.5 Evaluation Prediction
	4 Experiments and Analysis
		4.1 Dataset
		4.2 Experimental Setup
		4.3 Result Analysis
		4.4 Result Analysis
		4.5 Effect of Aggregation Functions on Recommended Performance
		4.6 Performance of the Model on Specific Datasets
	5 Conclusion
	Acknowledgment
	References
Improving Personalized Project Recommendation on GitHub Based on Deep Matrix Factorization
	1 Introduction
	2 Related Work
		2.1 GitHub Project Recommendations
		2.2 Deep Learning in Recommendation Systems
	3 Proposed Methods
		3.1 Data Collection
		3.2 Recommender System
		3.3 Result and Evaluation
	4 Experimental Setup
		4.1 Datasets
		4.2 Evaluation and Metrics
		4.3 Statistic Test
	5 Experimental Results
		5.1 RQ1:Does the Proposed Method Perform Better Than Other Comparison Methods?
		5.2 RQ2:What Is the Effect of the Dimension of the Low-Dimensional Vector and the Number of Recommended Lists on the Performance of the Proposed Method?
	6 Threats to Validity
		6.1 Internal Validity
		6.2 External Validity
	7 Conclusions
	References
An Intelligent SDN DDoS Detection Framework
	1 Introduction
	2 Related Works
	3 Security-oriented Flow Monitoring and Sampling
		3.1 Performance Analysis of Security-Oriented Flow Table Sampling
		3.2 Flow Monitoring and Sampling with Low-Latency Based on Optimization Theory
	4 Service Flow-Oriented Attack Recognition Model
		4.1 Service Flow Features Required by the Model
		4.2 DDoS Attack Detection Model Based on Clustering and VAE
		4.3 DDoS Attack Defense Based on Recognition Result
	5 Simulation and Performance Evaluation
		5.1 Simulation Setup
		5.2 Network Traffic Sampling Efficiency Evaluation
		5.3 Attack Detection Model Evaluation
	6 Conclusion
	References
Inspector: A Semantics-Driven Approach to Automatic Protocol Reverse Engineering
	1 Introduction
	2 Related Work
	3 System Design
		3.1 Overview
		3.2 Length Field Inference
		3.3 Message Type Field Inference
		3.4 Protocol Format Inference
	4 Evaluation
		4.1 Datasets
		4.2 Evaluation Metrics
		4.3 Tunable Parameters
		4.4 Experimental Results
	5 Conclusions
	References
MFF-AMD: Multivariate Feature Fusion for Android Malware Detection
	1 Introduction
	2 Related Work
	3 Multivariate Feature Extraction
		3.1 Static Feature Extraction
		3.2 Dynamic Feature Extraction
		3.3 Application Coverage
		3.4 Feature Selection
	4 Implementation
		4.1 Architecture
		4.2 Weight Distribution Algorithm
	5 Evaluation
		5.1 Dataset and Setup
		5.2 Results and Analysis
	6 Conclusion and Future Work
	References
Network and Security
PSG: Local Privacy Preserving Synthetic Social Graph Generation
	1 Introduction
	2 Related Work
		2.1 Social Network Privacy Protection
		2.2 Synthetic Graph Generation
	3 Preliminaries
		3.1 System Overview
		3.2 Problem Statement
	4 Design Details
		4.1 Privacy Protection Mechanism Design
		4.2 Privacy Analysis
		4.3 Synthetic Network Generation
	5 Performance Evaluation
		5.1 Datasets and Models
		5.2 Evaluation Metrics
		5.3 Experimental Results
	6 Conclusion
	References
Topology Self-optimization for Anti-tracking Network via Nodes Distributed Computing
	1 Introduction
	2 Related Works
	3 Introduction to Convex-Polytope Topology
		3.1 Basic Properties
		3.2 The Optimum Structure of CPT
	4 Topology Self-optimization
		4.1 Calculation of Optimum Local Topology
		4.2 Topology Self-optimization via Nodes’ Collaboration
	5 Performance Evaluation
		5.1 Evaluation of Network Optimization
		5.2 Evaluation of Network Resilience
	6 Conclusion
	References
An Empirical Study of Model-Agnostic Interpretation Technique for Just-in-Time Software Defect Prediction
	1 Introduction
	2 Related Work
		2.1 Just-in-time Software Defect Prediction
		2.2 Explainability in Software Defect Prediction
	3 Classifier-Agnostic Interpretation Technique
		3.1 LIME
		3.2 BreakDown
		3.3 SHAP
	4 Experimental Setup
		4.1 Data Sets
		4.2 Building Classification Models
		4.3 Evaluation Metrics
	5 Experimental Results and Analysis
		5.1 Analysis for RQ1
		5.2 Analysis for RQ2
	6 Conclusion and Future Work
	References
Yet Another Traffic Black Hole: Amplifying CDN Fetching Traffic with RangeFragAmp Attacks
	1 Introduction
	2 Background
		2.1 CDN Overview
		2.2 HTTP Range Request Mechanism on CDN
		2.3 Differences in CDNs Handling Range Requests
		2.4 Amplification Attacks
	3 RangeFragAmp Attack
		3.1 Threat Model
		3.2 S-RFA Attack
		3.3 O-RFA Attack
	4 Real-World Evaluation
		4.1 Consideration in Selecting CDN Providers
		4.2 S-RFA Attack Evaluation
		4.3 O-RFA Attack Evaluation
		4.4 Severity Assessment
		4.5 CDN Providers Feedback
	5 Mitigation
		5.1 Root Cause Analysis
		5.2 Limitation
		5.3 Solutions
	6 Related Work
	7 Conclusion
	References
DCNMF: Dynamic Community Discovery with Improved Convex-NMF in Temporal Networks
	1 Introduction
	2 Related Work
	3 Algorithm
		3.1 Notation
		3.2 The Unified DCNMF Model Formulation
		3.3 Optimization
	4 Experiments and Results
		4.1 Evaluation Measures
		4.2 Synthetic Dataset 1: Dynamic-GN Dataset
		4.3 Synthetic Dataset 2: Dynamic-LFR Dataset
		4.4 KIT-Email Data
	5 Discussion and Conclusion
	References
Network and Security and IoT and Social Networks
Loopster++: Termination Analysis for Multi-path Linear Loop
	1 Introduction
	2 Preliminaries
		2.1 Scope of Our Work
		2.2 Path Dependency Automaton (PDA)
		2.3 The Structure of Loopster
		2.4 Termination of Linear Loop Program
	3 Methodology
		3.1 Path Termination Analysis
		3.2 Inter-Path Analysis
		3.3 Cycle Analysis
	4 Implementation and Evaluation
		4.1 Effectiveness of Loopster++
		4.2 Performance of Loopster++
	5 Relate Work
	6 Conclusion
	References
A Stepwise Path Selection Scheme Based on Multiple QoS Parameters Evaluation in SDN
	1 Introduction
	2 Related Work
	3 Proposed Scheme: SWQoS
		3.1 SWQoS Scheme Architecture
		3.2 Path Finding
		3.3 QoS Requirements of Services
		3.4 Path Selection
	4 Experiments and Performance Evaluation
		4.1 The Experimental Environment and Topology
		4.2 The First Group Experiment: Simulating the Network Status of Selecting the Preferred Paths
		4.3 The Second Group Experiment: Simulating the Network Status of Obtaining Satisfied Paths
		4.4 The Third Group Experiment: Simulating the Network Status of Obtaining Reluctant Paths
	5 Conclusions
	References
A Novel Approach to Taxi-GPS-Trace-Aware Bus Network Planning
	1 Introduction
	2 Related Work
	3 Main Steps
		3.1 Candidate Bus Stop Identification
		3.2 Bus Network Generation
	4 Simulations
	5 Conclusion
	References
Community Influence Maximization Based
on Flexible Budget in Social Networks
	1 Introduction
	2 Related Work
	3 System Model and Problem Formulation
		3.1 System Model
		3.2 Problem Formulation
	4 Our Solutions
		4.1 General Solution
		4.2 FBCIM Algorithm
		4.3 FBBCIM Algorithm
	5 Performance Evaluation
		5.1 Datasets and Parameters Setting
		5.2 Comparison of Algorithms and Metrics
		5.3 Evaluation Results
	6 Conclusion
	References
An Online Truthful Auction for IoT Data Trading with Dynamic Data Owners
	1 Introduction
	2 System Model and Problem Formulation
		2.1 System Model
		2.2 Data Trading Model Based on an Auction Mechanism
		2.3 Problem Formulation
	3 Online Data Trading Algorithm
		3.1 Online Matching Algorithm Based on a Greedy Strategy
		3.2 Computing Trading Prices Based on Critical Data Owners
		3.3 Theoretical Analysis
	4 Numerical Illustration
		4.1 Methodology and Simulation Settings
		4.2 Numerical Results
	5 Related Work
		5.1 Decentralized Data Trading Based on the Blockchain Technology
		5.2 Trading Data with Different Levels of Privacy
	6 Conclusion
	References
IoT and Social Networks and Images Handling and Human Recognition
Exploiting Heterogeneous Information for IoT Device Identification Using Graph Convolutional Network
	1 Introduction
	2 Preliminaries
		2.1 TLS Basics
		2.2 Graph Convolutional Networks
		2.3 Problem Definition
	3 The THG-IoT Framework
		3.1 Data Preprocessing
		3.2 Graph Generation
		3.3 GCN Classifier
	4 Experimental Evaluation
		4.1 Dataset
		4.2 Evaluation Metrics
		4.3 Experimental Setting
		4.4 Parameter Study
		4.5 Comparison Experiments
		4.6 Variant of THG-IoT
	5 Related Work
	6 Conclusions and Future Work
	References
Data-Driven Influential Nodes Identification in Dynamic Social Networks
	1 Introduction
	2 Related Work
	3 Data-Driven Model for Influential Nodes Identification in Social Networks
		3.1 Multi-scale Comprehensive Metric System
		3.2 Data-Driven Weight Optimization Algorithm
		3.3 Influential Nodes Identification Based on Data-Driven Weighted TOPSIS
	4 Experiments and Analysis
		4.1 Experimental Setup
		4.2 Performance Comparison
	5 Conclusion and Future Work
	References
Human Motion Recognition Based on Wi-Fi Imaging
	Abstract
	1 Introduction
	2 Wi-Fi Imaging Algorithm Based on 3D Virtual Array
		2.1 Imaging Algorithm Based on Virtual 3D Array
		2.2 Description of Improved 3D Decoherence Algorithm
	3 Environment Adaptive Human Continuous Motion Recognition
		3.1 Continuous Action Segmentation
		3.2 Action Feature Extraction
		3.3 SVM Classification Based on GA Algorithm Optimization
	4 Experiment and Result Analysis
		4.1 Experimental Configuration
		4.2 Human Imaging and Motion Recognition
		4.3 Result Analysis
		4.4 Model Test
		4.5 Comparison of Different Models
	5 Conclusion
	Acknowledgment
	References
A Pervasive Multi-physiological Signal-Based Emotion Classification with Shapelet Transformation and Decision Fusion
	Abstract
	1 Introduction
	2 Related Works
		2.1 Emotion Classification Based on Physiological Signals
		2.2 Shapelet-Based Algorithms
	3 Methods
		3.1 Overview
		3.2 Data Preprocessing
		3.3 Sub-classification Methods
			3.3.1 Shapelet Transformation Algorithm
			3.3.2 Feature Extraction
			3.3.3 Sub-classifiers
		3.4 Decision-Level Fusion Strategy
	4 Experimental Results and Analysis
		4.1 Database
		4.2 Results of Emotion Classification of a Single Physiological Signal
		4.3 Results Comparisons
	5 Conclusion and Future Work
	Acknowledgments
	References
A Novel and Efficient Distance Detection Based on Monocular Images for Grasp and Handover
	1 Introduction
	2 Related Works
		2.1 RGB-D-Based Methods
		2.2 Analytic-Based Methods
		2.3 Model-Based Methods
	3 Method
		3.1 Distance Detection A
		3.2 Distance Detection B
	4 Experiments and Results
		4.1 Experimental Equipment
		4.2 Preliminary Work
		4.3 Grasping Tests
		4.4 Human-Robot Handover Tests
		4.5 Time Cost
		4.6 Qualitative Results and Future Work
	5 Conclusion
	References
Images Handling and Human Recognition and Edge Computing
A Novel Gaze-Point-Driven HRI Framework for Single-Person
	1 Introduction
	2 Related Work
		2.1 Gaze Point Estimation
		2.2 Application of Gaze Points in HRI
	3 Methods
		3.1 Overview
		3.2 Object Locations Distribution Obtaining
		3.3 Gaze Points Distribution Estimating
		3.4 Gaze Target Reasoning and Entity Matching
		3.5 Moving and Grabbing
	4 Results
		4.1 Experimental Equipment and Setup
		4.2 Experimental Results
	5 Conclusion
	References
Semi-automatic Segmentation of Tissue Regions in Digital Histopathological Image
	1 Introduction
	2 Related Work
		2.1 Methods Based on Hand-Crafted Features
		2.2 Methods Based on Deep Learning
	3 Preliminaries
		3.1 Methodology Overview
		3.2 Histopathological Images Preprocessing: Staining Normalization
		3.3 Pre-segmentation of Tissue Regions
		3.4 Automatic Segmentation of Tissue Regions
	4 Experiments and Results Analysis
		4.1 Experimental Objective
		4.2 Dataset
		4.3 Experimental Setup
		4.4 Experimental Results and Analysis
	5 Conclusion and Future Work
	References
T-UNet: A Novel TC-Based Point Cloud Super-Resolution Model for Mechanical LiDAR
	1 Introduction
	2 Related Works
	3 Model Architecture
		3.1 Point Cloud Projection and Back-Projection
		3.2 T-UNet Model
	4 Experimental Study
		4.1 Datasets
		4.2 Implementation Details
		4.3 Model Evaluation
	5 Conclusion
	References
Computation Offloading for Multi-user Sequential Tasks in Heterogeneous Mobile Edge Computing
	1 Introduction
	2 Related Work
	3 MUST Model and Problem Formulation
		3.1 System Model
		3.2 Problem Formulation
	4 Regular Expression Based Algorithm for MUST
	5 Performance Evaluation
		5.1 Setup
		5.2 Results
	6 Conclusion
	References
Model-Based Evaluation and Optimization of Dependability for Edge Computing Systems
	1 Introduction
	2 Related Work
	3 Dependability Modeling and Analysis of Edge/Cloud Server
		3.1 System State Transition Model
		3.2 Analysis of Dependability Attributes
	4 Dependability Modeling and Analysis of Edge Computing System
		4.1 State Aggregation Technique
		4.2 Dependability Model of Edge Computing Systems
	5 Model and Approach of Dependability Optimization
	6 Empirical Evaluation
		6.1 Data Set and Experimental Settings
		6.2 Experimental Results
	7 Conclusion
	References
Author Index