Computational Data and Social Networks: 12th International Conference, CSoNet 2023, Hanoi, Vietnam, December 11–13, 2023, Proceedings (Lecture Notes in Computer Science, 14479)

This book constitutes the refereed conference proceedings of the 12th International Conference on Computational Data and Social Networks, CSoNet 2023, held in Hanoi, Vietnam, in December 2023. The 23 full papers and 14 short papers presented in this book were carefully reviewed and selected from 64 submissions. The papers are divided into the following topical sections: machine learning and prediction; optimization; security and blockchain; and network analysis. Preface Organization Contents Machine Learning and Prediction An Approach Using Threshold-Based Noise Reduction and Fine-Tuned ShuffleNetV2 for Plant Leaf Disease Detection 1 Introduction 2 Related Work 3 Methods 3.1 Dataset 3.2 Soft-Thresholding with Lasso Regression for Noise Reduction 3.3 Deep Learning Architectures for Disease Detection on Leaf Images 4 Experimental Results 4.1 Experimental Setup 4.2 Scenario 1: Disease Performance with ShuffleNetV2 4.3 Scenario 2: Disease Performance with ResNet50 4.4 Scenario 3: Disease Performance with LeNet 4.5 Scenario 4: Soft-Thresholding with Lasso Regression 4.6 Discussion 5 Conclusion References Enhancing Visual Question Answering with Generated Image Caption 1 Introduction 2 Related Work 2.1 Visual Question Answering 2.2 VQA with Image Captioning 3 Methodology 3.1 Caption Generation 3.2 Feature Representation 3.3 Multimodal Representation Aggregation 3.4 Answers Decoder 4 Experiments 4.1 Dataset 4.2 Metrics 4.3 Implementation Details 4.4 Results and Discussions 5 Conclusion References A Gaussian Distribution Labeling Method for Speech Quality Assessment 1 Introduction 2 Proposed Method 2.1 Gaussian Distribution Labeling 2.2 Focal Loss 2.3 Model Architecture 3 Experiments 3.1 Setup 3.2 Datasets 3.3 Scenario 3.4 Results and Discussion 4 Conclusions References TextFocus: Efficient Multi-scale Detection for Arbitrary Scene Text 1 Introduction 2 Related Works 3 Methodology 3.1 Detection Branch 3.2 Focus Branch 3.3 Combination for Final Results 4 Experiments and Results 4.1 Datasets 4.2 Experiments 4.3 Result and Analysis 5 Conclusion References Fake Face Recognition on Images Generated by Various Deepfakes Tools 1 Introduction 2 Related Work 3 Methods 3.1 Overall Workflow 3.2 Tools for Fake Face Generation 3.3 The Convolutional Neural Network Architecture 3.4 Dataset 4 Experimental Results 4.1 Environmental Settings and Metrics 4.2 Scenario 1: The Ability to Make Fake Faces by the Considered Tools 4.3 Scenario 2: The Comparison Between the Proposed Architecture and MobiletNet 5 Conclusion References Multi-scale Aggregation Network for Speech Emotion Recognition 1 Introduction 2 Proposed Methodology 2.1 Multi-filter Convolutional Neural Network (MFCNN) 2.2 Multi-scale Feature Aggregation (MSA) Network 2.3 Data Augmentation Methods 2.4 Loss Function 3 Experiments 3.1 Datasets 3.2 Experimental Results and Analysis 3.3 Conclusions References ViEcomRec: A Dataset for Recommendation in Vietnamese E-Commerce 1 Introduction 2 Related Work 3 Dataset 3.1 Dataset Crawling 3.2 Attribute Extraction 4 Methodology 4.1 Problem Definition 4.2 Content-Based Filtering with Popularity Score 4.3 Attribute-Based Filtering 5 Experiment 5.1 Experimental Settings 5.2 Experimental Results 6 Conclusions and Future Work References Untargeted Code Authorship Evasion with Seq2Seq Transformation 1 Introduction 2 Our Approach: Building Blocks 3 SCAE: Putting it Together 4 Evaluation 4.1 Experimental Setup and Goal 4.2 Results and Analysis 5 Conclusion References Mining the Potential Temporal Features Based on Wearable EEG Signals for Driving State Analysis 1 Introduction 2 EEG-Based Fatigue Driving Detection Model 2.1 Feature Extraction 2.2 Model Training and Optimization 3 The Experimental Analysis 4 Conclusion References VN-Legal-KG: Vietnam Legal Knowledge Graph for Legal Statute Identification on Land Law Matters 1 Introduction 2 Vietnam Legal Knowledge Graph 3 Link Prediction for Multi-label LSI Prediction 3.1 GAT Training with VN-Legal-KG 3.2 Link Prediction for a New Question 4 Experiments 4.1 Dataset 4.2 Training and Evaluating Protocol 4.3 Baselines 5 Conclusion References CombiGCN: An Effective GCN Model for Recommender System 1 Introduction 2 Related Work 2.1 Graph Convolution Networks 2.2 Multi-layer Perceptron and Backpropagation 3 Our Proposed Method 3.1 Pre-processing Data 3.2 Adjacent Graph and Weighted References Matrix 3.3 CombiGCN 4 Experiments 4.1 Datasets Description 4.2 Experimental Settings 4.3 Experiment Results 5 Conclusion References Unveiling Sentiments in Vietnamese Education Texts: Could Large Language Model GPT-3.5-turbo Beat PhoBERT? 1 Introduction 2 Related Works 3 Proposed Model 3.1 PhoBERT Pre-trained Model and Fine-Tuning for Specified Task 3.2 Other Tested Models Including Variants of BERT and Large Language Model (GPT-3.5) 4 Result 4.1 Dataset 4.2 Experimental Result 5 Conclusion References Portfolio Construction Based on Time Series Clustering Method Evidence in the Vietnamese Stock Market 1 Introduction 2 Data and Methodology 2.1 Data 2.2 Methodology 3 Empirical Analysis and Results 4 Conclusion References An Approach for Web Content Classification with FastText 1 Introduction 2 Related Works 3 Methods 3.1 Data Preprocessing 3.2 Text Classification 4 Experiments 4.1 Dataset 4.2 Scenario 1: Text Classification with TF-IDF 4.3 Scenario 2: Text Classification with TF-IDF with SVD 4.4 Scenario 3: Text Classification with FastText 4.5 Evaluate the Experimental Results 5 Conclusion References Optimization Continuous Length-Bounded Paths Interdiction 1 Introduction 2 Preliminaries 2.1 Problem Formulation 2.2 General Model of Our Solutions 3 Threshold Blocking Oracle 3.1 Requirements of TB 3.2 Threshold Expansion 4 Critical Path Listing Oracle 5 Conclusion References A* Search Algorithm for an Optimal Investment Problem in Vehicle-Sharing Systems 1 Introduction 2 Literature 3 Problem Statement and Formulation 3.1 Semi-Markov Decision Process for Determining the Optimal Fleet Size 3.2 A Model for Calculating Optimal Profit and Minimum Acquisition Cost 4 Solution Approach 4.1 Heuristic Strategy for A* Algorithm 5 Numerical Experiments 5.1 Numerical Design 5.2 Results 6 Conclusion References The Matheuristic for Building Cost-Constrained Decision Trees with Multiple Condition Attributes 1 Introduction 2 Literature 3 Problem Statement and Existing Formulation 4 The Matheuristic Approach 4.1 The MILP Formulation for Assigning Decision Attributes 4.2 The MILP-Based Algorithm: VND 5 Numerical Experiments 6 Conclusion References Fast Bicriteria Approximation Algorithm for Minimum Cost Submodular Cover Problem 1 Introduction 2 Related Works 3 Preliminaries 4 Proposed Algorithms for the Problem 4.1 Proposed Algorithm with Known opt 4.2 Main Algorithm 5 Empirical Evaluation 6 Conclusion References Mixed Integer Linear Programming-Based Methods for the Optimal Time-Constrained Cost-Sensitive Decision Tree 1 Introduction 2 Mathematical Formulation 3 The Adaptation of Limited Times and Costs of Features in Existing Models 4 Numerical Experiments 5 Conclusion References Solving Time-Dependent Traveling Salesman Problem with Time Windows Under Generic Time-Dependent Travel Cost 1 Problem Formulation 2 Literature Review 3 Partially Time-Expanded Network Formulation and Properties 4 New Lower Bound Formulations and Algorithm 4.1 Path-arc-based Formulation 4.2 Arc-Based Formulation 4.3 Algorithm to Solve TD-TSPTW(DT) 5 Experiments 6 Conclusion References Improved Streaming Algorithm for Minimum Cost Submodular Cover Problem 1 Introduction 2 Related Works 3 Preliminaries 4 Proposed Algorithm 5 Experimental Evaluation 5.1 Applications and Datasets 5.2 Experiment Results 6 Conclusion References Deletion-Robust Submodular Maximization Under the Cardinality Constraint over the Integer Lattice 1 Introduction 2 Preliminaries 3 Robustness Setting over the Integer Lattice 3.1 Centralized Algorithm over the Integer Lattice 3.2 Streaming Setting over the Integer Lattice 4 Conclusion References Evolutionary Parameter Optimization: A Novel Control Strategy for Chaotic Environments 1 Introduction 2 Approach and Execution 2.1 Trend Identification Strategies 2.2 Evolutionary Modeling for Optimal Control Strategies: A Profits-Based Approach 3 Experimental Procedures and Findings 3.1 Investigating Suboptimal Initial Threshold Values: Estimation and Analysis 3.2 Implementation Details 3.3 Evolutionary Optimization in Control Strategy: Application, Modifications, and Parameter Estimation 4 Conclusion References Security and Blockchain The Infrastructure Utilization of Free Contents Websites Reveal Their Security Characteristics 1 Introduction 2 Related Work 3 Methodology 3.1 Research Questions 3.2 Data Collection Process 3.3 Correlation Analysis 4 Correlation Results 4.1 General Correlation Results 4.2 Networks Correlations 4.3 Cloud Service Providers 4.4 Countries Correlation 5 Results Discussion 5.1 Results Takeaway 5.2 Limitations and Recommendations 6 Conclusion References Understanding the Utilization of Cryptocurrency in the Metaverse and Security Implications 1 Introduction 2 Related Work 3 Problem Statement and Research Questions 4 Technical Approach 4.1 Dataset and Preprocessing 4.2 Analysis Dimensions 5 Results and Findings 5.1 Communication and Referrer Files in the Metaverse Domain 5.2 Metaverse Coins Market Capitalization 5.3 Malicious Activities in Metaverse Coins 5.4 Metaverse Coins with Fiat Currency to Cryptocurrency 6 Discussion 7 Conclusion and Future Work References Blockchain and IoT for Enhanced Traceability in Waste Treatment Processes: A Microservice and Brokerless Approach 1 Introduction 2 Related Work 3 Approach 3.1 Traditional Monitoring Approach 3.2 Automated Control Approach 3.3 Blockchain and IoT-Based Approach 4 Evaluation 4.1 Evironment Setting 4.2 Microservice Architecture Test Case 4.3 Hyperledger Caliper Test Case 5 Discussion 6 Conclusion References Towards a Management System of Blood and Its Products Based on Blockchain, Smart Contracts, and NFT: A Case Study in Vietnam 1 Introduction 2 Related Work 3 Approach 3.1 Traditional Model of Blood Donation and Blood Management 3.2 Blood Donation and Blood Management Model Based on Blockchain Technology, Smart Contracts, and NFTs 4 Evaluation Scenarios 4.1 Transaction Fee 4.2 Gas Limit 4.3 Gas Price 5 Conclusion References Secure Conversation: A View from Physical Layer 1 Introduction 2 System Model and Objective Function 2.1 System Model 2.2 Performance Analysis 3 Numerical Results 4 Conclusions References MIDAS: A Multi-chain Interoperable Data and Access System for Healthcare 1 Introduction 2 Related Works 2.1 Cross-Chain Technology 2.2 Identity Management Systems and Decentralized Identifiers 3 Proposed System 3.1 BIDI - Blockchain Interoperability Decentralized Identifier 3.2 MIDAS - Multi-chain Interoperable Data and Access System 3.3 Data Authorization Mechanism 4 Experiment Results 5 Conclusion References Network Analysis Leveraging GNNs and Node Entropy for Anomaly Detection: Revealing Misinformation Spreader on Twitter Network 1 Introduction 2 Related Literature 3 Theoretical Framework and Approaches 3.1 Graph Neural Networks (GNNs) and Node Embeddings 3.2 Node Entropy 4 Study Framework and Investigative Methodology 4.1 Dataset 4.2 Experimentation and Result 4.3 Methodological Soundness and Justification 5 Discussion and Analysis 6 Conclusion and Future Work References Air Transportation Network Backbone Extraction: A Comparative Analysis of Structural Filtering Techniques 1 Introduction 2 Structural Backbone Extraction Methods 3 Data, Evaluation Measures, and Methods 3.1 Data 3.2 Evaluation Measures 3.3 Methods 4 Experimental Results 4.1 Investigating the Backbones Similarity 4.2 Investigating the Relation Between Backbone Edges and Weights 4.3 Comparing Backbones Basic Properties 4.4 Comparing Backbones Connectivity and Transitivity 5 Discussion 6 Conclusion References Multiplex Network Approach for Modeling the Spread of African Swine Fever in Poland 1 Introduction 1.1 ASF Epidemiology 2 Data Preparation, Methodology and Descriptive Analysis 2.1 Regressions 2.2 Network Representation 3 Results 3.1 Simulation Scenario: Border Fencing 3.2 Simulation Scenario: A1 Blocking 3.3 Arrival Times Without Intervention Before and After Wschowski Case 4 Discussion and Conclusions References Identification and Analysis of the Spread of {Mis}information on Social Media 1 Introduction 2 Background 3 Methods and Experimental Design 4 Results and Analysis 5 Related Work 6 Discussion 7 Conclusion References Modeling the Change of Public Opinion Caused by Opinion Posts' Dissemination in Social Networks 1 Introduction 2 Model Design 2.1 Model's Parameters and Individuals' Attributes 2.2 Opinion Post Propagation 2.3 Individual's Opinion Update 2.4 Model Framework 3 Simulation 4 Conclusion References Scalable Deep Metric Learning on Attributed Graphs 1 Introduction 2 Preliminaries 3 Methodology 3.1 Problem Statement 3.2 DMT Algorithm 3.3 DMAT-i Algorithm 4 Theoretical Analysis 5 Experimental Results 5.1 Results on Downstream Tasks 6 Conclusions References Deep Learning Based Model for Stress Measurement in Online Social Networks 1 Introduction 2 Related Work 3 Proposed Model 4 Results and Discussion 4.1 Data Preprocessing 4.2 Accuracy and Loss Curves 4.3 Classification Report 4.4 Confusion Matrix 5 Conclusion References Stress Expression Identification Model for Emotion-Driven Association Calculation over Social Network Blogs 1 Introduction 2 Emotion Sequence Pattern Analysis 2.1 Stress Emotions Dictionary Construction 2.2 Stress Emotion Sequence Intensity Quantification 3 Stress Expression Identification Model 4 Experiment 4.1 Classification Algorithms Evaluation 4.2 SEI Model Evaluation 4.3 Actual Effect Evaluation 5 Conclusion References Extended Abstracts Analytics for an E-commerce: DemandAnalysis and Prediction Maximizing Regularized DR-SubmodularFunctions on a Convex Set References Activation and Deactivation Phenomenain Online Social Media Network CascadePropagation with Geographic Considerations Reference The Impact of News, Expert and Public Opinionon Painting Prices 1 Introduction 2 Method 3 Results References Author Index