Intelligent Information and Database Systems: 14th Asian Conference, ACIIDS 2022, Ho Chi Minh City, Vietnam, November 28–30, 2022, Proceedings, Part I

Description

This book constitutes the refereed proceedings of the 14th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2022, held Ho Chi Minh City, Vietnam in November 2022. The 113 full papers accepted for publication in these proceedings were carefully reviewed and selected from 406 submissions. The papers of the 2 volume-set are organized in the following topical sections: data mining and machine learning methods, advanced data mining techniques and applications, intelligent and contextual systems, natural language processing, network systems and applications, computational imaging and vision, decision support and control systems, and data modeling and processing for industry 4.0. The accepted and presented papers focus on new trends and challenges facing the intelligent information and database systems community. Preface Organization Contents – Part I Contents – Part II Advanced Data Mining Techniques and Applications Textual One-Pass Stream Clustering with Automated Distance Threshold Adaption 1 Introduction 1.1 Recent Work on Textual Stream Clustering 2 Distance Based Clustering with Automatic Threshold Determination (textClust) 2.1 Automatic Tresholding During the Online Phase 2.2 Algorithm Specification 3 Experiments 3.1 Benchmarking Datasets 3.2 Experimental Setup 3.3 Evaluation Metrics 3.4 Experimental Results 4 Discussion and Future Work References Using GPUs to Speed Up Genetic-Fuzzy Data Mining with Evaluation on All Large Itemsets 1 Introduction 2 Related Work 3 Components of the Proposed Algorithm 3.1 Chromosome Representation 3.2 Population Initialization 3.3 Fitness Function and Selection 3.4 Genetic Operators and Termination 4 The Proposed GFM-GPU-LAll Optimization Algorithm 5 Experimental Evaluations 6 Conclusions and Future Work References Efficient Classification with Counterfactual Reasoning and Active Learning 1 Introduction 2 Related Works 3 Framework 3.1 Problem Definition 3.2 Proposed Method CCRAL 4 Experiments and Discussions 4.1 Datasets 4.2 Baselines and Evaluation 4.3 Results 5 Conclusion References Visual Localization Based on Deep Learning - Take Southern Branch of the National Palace Museum for Example 1 Introduction 2 Related Work 2.1 Convolutional Neural Network 2.2 Visual Localization Based on Deep Learning 3 Proposed Method 3.1 Network Architecture 3.2 Loss Function 4 Experiments 4.1 Pretrained Model 4.2 Normalization 4.3 Loss Function 5 Conclusion and Future Work References SimCPSR: Simple Contrastive Learning for Paper Submission Recommendation System 1 Introduction 2 Related Work 3 Methodology 3.1 Contrastive Learning 3.2 Modeling 3.3 Evaluation Metrics 4 Experiments 4.1 Experimental Settings 4.2 Datasets 4.3 Training Details 4.4 Results 5 Conclusion and Further Works References Frequent Closed Subgraph Mining: A Multi-thread Approach 1 Introduction 2 Related Work 3 Definitions 4 Proposed Method 5 Experimental Results 6 Conclusion and Future Work References Decision Support and Control Systems Complement Naive Bayes Classifier for Sentiment Analysis of Internet Movie Database 1 Introduction 2 Related Work 2.1 Sentiment Analysis (SA) 2.2 Complement Naïve Bayes Classifier 2.3 Analysis Metrics 3 Methodology 3.1 Research Workflow 3.2 Internet Movie Database (IMDB) 4 Experiment and Result 4.1 Experiment Results 5 Conclusions References Portfolio Investments in the Forex Market 1 Introduction 2 Related Works 3 Proposed Methodology 3.1 The Investing Process 4 Numerical Experiments 5 Conclusions References Detecting True and Declarative Facial Emotions by Changes in Nonlinear Dynamics of Eye Movements 1 Introduction 2 Methods 3 Results 4 Discussions 5 Conclusions References Impact of Radiomap Interpolation on Accuracy of Fingerprinting Algorithms 1 Introduction 2 Related Work 2.1 Fingerprinting Localization 2.2 Dynamic Radiomap 2.3 Interpolation Algorithms 3 Experimental Scenario and Achieved Results 4 Conclusions References Rough Set Rules (RSR) Predominantly Based on Cognitive Tests Can Predict Alzheimer’s Related Dementia 1 Introduction 2 Methods 2.1 Theoretical Basis 3 Results 3.1 Statistical Results 3.2 RSR for Reference of Model1 Group 3.3 RSR for Reference of Model2 Group 4 Discussion References Experiments with Solving Mountain Car Problem Using State Discretization and Q-Learning 1 Introduction 2 Related Works 3 Modeling the Mountain Car Problem 3.1 Physics of the Mountain Car Problem 3.2 Model Exploration Using Random Walk and Numerical Simulation 4 Optimal Control Using State Discretization and Q-Learning 4.1 Q-Learning and SARSA Algorithms 4.2 State Discretization 4.3 Experimental Results 5 Conclusions and Future Work References A Stable Method for Detecting Driver Maneuvers Using a Rule Classifier 1 Introduction 2 Data Logging 2.1 Data Stream Forming 2.2 Data Collection 3 Evaluation of the Model 4 Conclusions and Further Work References Deep Learning Models Using Deep Transformer Based Models to Predict Ozone Levels 1 Introduction 2 Related Work 3 Preliminaries 3.1 Baseline Models 3.2 Performance Evaluation Metrics 4 Problem Description and Our Model 4.1 Problem Description 4.2 Deep Transformer Based Models 4.3 MLP and LSTM Networks 5 Experiments 5.1 Comparison Between Models 5.2 Hyperparameters Optimisation 6 Conclusions and Future Work References An Ensemble Based Deep Learning Framework to Detect and Deceive XSS and SQL Injection Attacks 1 Introduction 1.1 Background Study 2 Proposed Detection and Deception Technique 2.1 Data Preparation and Feature Selection 2.2 Using the Ensemble Based Deep Learning Classifiers 2.3 State Maintenance Module 2.4 Deception Module to Lure/Engage Attackers 3 Discussion, Performance Analysis and Testing 3.1 Comparative Analysis 4 Conclusion and Future Work References An Image Pixel Interval Power (IPIP) Method Using Deep Learning Classification Models 1 Introduction 2 Related Works 3 Proposed Methodology 4 Experiments and Results 4.1 Datasets 4.2 Baseline Method 4.3 Training Setup 4.4 Evaluation Metrics 4.5 Experimental Results and Discussions 5 Conclusion References Meta-learning and Personalization Layer in Federated Learning 1 Introduction 2 Related Work 3 Proposed Method 4 Numerical Experiments 5 Results and Discussion 6 Conclusion A Experimental Details A.1 Model Architecture A.2 Hyper-parameters Searching References ETop3PPE: EPOCh’s Top-Three Prediction Probability Ensemble Method for Deep Learning Classification Models 1 Introduction 2 Related Works 3 Proposed Method 4 Experiments and Results 4.1 Dataset 4.2 Training Setup 4.3 Evaluation Metrics 4.4 Experiment Results and Discussions 5 Conclusions References Embedding Model with Attention over Convolution Kernels and Dynamic Mapping Matrix for Link Prediction 1 Introduction 2 Related Work 3 Background 3.1 Dynamic Convolution 3.2 TransD Model 4 The Proposed Model 5 Experiments and Result Analysis 5.1 Benchmark Datasets 5.2 Experimental Setup 5.3 Results 6 Conclusion and Future Research Directions References Employing Generative Adversarial Network in COVID-19 Diagnosis 1 Introduction 2 Proposed Framework 2.1 Data Augmentation 2.2 Transfer Learning 3 Experimental Evaluation 3.1 Using GAN to Generate Synthetic Images 3.2 Transfer Learning 4 Conclusion References SDG-Meter: A Deep Learning Based Tool for Automatic Text Classification of the Sustainable Development Goals 1 Introduction 2 State-of-the-Art 3 Multi-labeled Text Classification with BERT 3.1 BERT: Bidirectional Encoder Representations from Transformers 3.2 SDG-Meter Tool 4 Experimentation 4.1 Dataset 4.2 Test and Results 5 Conclusion References The Combination of Background Subtraction and Convolutional Neural Network for Product Recognition 1 Introduction 2 Related Work 3 Methodology 3.1 Background Subtraction and Skin Removal 3.2 Product Classification 3.3 Product Tracking and Counting 4 Experiment 4.1 Experimental Setup 4.2 Training Classifier 4.3 Result and Discussion 5 Conclusions References Strategy and Feasibility Study for the Construction of High Resolution Images Adversarial Against Convolutional Neural Networks 1 Introduction 1.1 Attacks in the R Domain 1.2 Three Challenges Faced by Attacks in the H Domain 1.3 Our Contribution: A Strategy and a Feasibility Study 2 CNNs and the Target Scenario 2.1 The Target Scenario 2.2 The Target Scenario Lifted to HR Images 3 Attack Strategy for the Target Scenario on HR Images 3.1 Construction of Adversarial Images in H 3.2 Indicators: The Loss Function L and L2-distances 4 Feasibility Study 4.1 The Evolutionary Algorithm EAtarget,C 4.2 Running the Strategy to Get Adversarial Images with the EA 4.3 Visual Quality 5 Conclusion References Using Deep Learning to Detect Anomalies in Traffic Flow 1 Introduction 2 Problem Description 2.1 Data 2.2 Scenarios 3 Auto-encoder Models 3.1 CNN Auto-encoder Model 3.2 BiLSTM Auto-encoder Model 4 Experiments 4.1 Basic Scenario 4.2 Guided Scenario 5 Conclusions and Future Work References A Deep Convolution Generative Adversarial Network for the Production of Images of Human Faces 1 Introduction 2 A Recall of the Genarative Adversial Networks (GAN) 3 Related Works Concerning the Variants of GANs 3.1 Architecture-Variant 3.2 Loss-Variant 4 Deep Convolutional GAN: A Method Adopted for Human Faces Images Producing 4.1 Datasets 4.2 Configuration 4.3 Results 5 Evaluation 5.1 Our Contribution 5.2 Evaluation of the Images Produced by DCGAN with MEQFI 6 Conclusion References ECG Signal Classification Using Recurrence Plot-Based Approach and Deep Learning for Arrhythmia Prediction 1 Introduction 2 Data Set 3 Method 3.1 Time Series to Recurrent Plots 3.2 Convolutional Neural Networks 3.3 Fine Tuning CNNs 3.4 Classification Performance Evaluation 4 Results and Discussion 5 Conclusion and Future Works References Internet of Things and Sensor Networks Collaborative Intrusion Detection System for Internet of Things Using Distributed Ledger Technology: A Survey on Challenges and Opportunities 1 Introduction 1.1 Related Surveys on Collaborative Intrusion Detection Systems for IoT 1.2 Overview of the Paper 2 Distributed Ledger Technology 3 Collaborative Intrusion Detection Systems Based on Distributed Ledger Technology in Internet of Things 3.1 CIDS Placement Strategies 3.2 Detection Method 3.3 Security Threat 3.4 Validation and Testing Method 3.5 DLT Platform 4 Future and Opportunities 4.1 Challenge in Research 4.2 Future Research and Opportunities 5 Conclusion References An Implementation of Depth-First and Breadth-First Search Algorithms for Tip Selection in IOTA Distributed Ledger 1 Introduction 2 Related Works 3 Problem Description 3.1 Breadth-First Search 3.2 IRI Implementation of Cumulative Weight Calculation 4 Proposed Solutions 4.1 Depth-First Search (DFS) 4.2 Our Solution 5 Experimental Results 5.1 Sample Implementation 5.2 Experiments and Discussion 6 Conclusion and Future Works References Locally Differentially Private Quantile Summary Aggregation in Wireless Sensor Networks 1 Introduction 2 Related Work 3 Problem Formulation 3.1 Network Model 3.2 Quantile Summary 3.3 Local Differential Privacy (LDP) 3.4 Design Goals 4 PrivQSA: Quantile Summary Aggregation with LDP 4.1 Overview 4.2 Detailed Design 5 Performance Evaluation 5.1 Theoretical Analysis 5.2 Simulation Settings 5.3 Simulation Results 6 Conclusion References XLMRQA: Open-Domain Question Answering on Vietnamese Wikipedia-Based Textual Knowledge Source 1 Introduction 2 Related Work 3 UIT-ViQuAD: Vietnamese Wikipedia-Based Textual Knowledge Resource 4 XMLRQA: Retriever-Reader-Selector Question Answering System for the Vietnamese Language 4.1 Overview of QA System Architecture 4.2 Text Retriever 4.3 Text Reader 4.4 Answer Selector 5 Experimental Evaluation 5.1 Baseline Systems 5.2 Experimental Settings 5.3 Experimental Results 5.4 Result Analysis 6 Conclusion and Future Work References On Verified Automated Reasoning in Propositional Logic 1 Introduction 2 Related Work 3 Sequent Calculus for Propositional Logic 4 Programming the Prover 5 Verifying the Prover 6 Using the Prover 7 Concluding Remarks References Embedding and Integrating Literals to the HypER Model for Link Prediction on Knowledge Graphs 1 Introduction 2 Related Work 3 The Proposed Method 4 Experiments and Result Analysis 4.1 Datasets 4.2 Metrics 4.3 Hyperparameters and Experimental Setup 4.4 Results 4.5 The Influence of Hyperparameters 5 Conclusion References A Semantic-Based Approach for Keyphrase Extraction from Vietnamese Documents Using Thematic Vector 1 Introduction 2 Related Work 3 Semantic-Based Approach for Keyphrase Exaction (SAKE) 3.1 Step 1: Preprocessing Documents 3.2 Step 2: Selecting Candidates 3.3 Step 3: Scoring Candidates 4 Experiments 4.1 Dataset 4.2 Evaluation 4.3 Results 5 Discussions 6 Conclusions References Mixed Multi-relational Representation Learning for Low-Dimensional Knowledge Graph Embedding 1 Introduction 2 Related Works 2.1 Euclidean Embeddings 2.2 Complex Embeddings 2.3 Hyperbolic Embeddings 3 Background 3.1 Riemannian Manifolds 3.2 Constant-Curvature Spaces 3.3 Problem Formulation 3.4 Lorentzian Distance 4 Methodology 4.1 MuREL 4.2 Training and Optimization 5 Experiments 5.1 Datasets 5.2 Evaluation Metrics 5.3 Settings and Hyperparameters 5.4 Low-Dimensional Embedding Results 5.5 High-Dimensional Embedding Results 6 Conclusion and Future Research Directions References Learning to Map the GDPR to Logic Representation on DAPRECO-KB 1 Introduction 2 Related Work 3 Methodology 3.1 Baseline NMT Model 3.2 Sub-expression Intersection Mechanism 3.3 PRESEG Mechanism 4 Experiments 4.1 Datasets 4.2 Experimental Settings 4.3 Experimental Results and Discussion 5 Conclusion and Future Work References Semantic Relationship-Based Image Retrieval Using KD-Tree Structure 1 Introduction 2 Related Works 3 Object Detection Using R-CNN Network Model 3.1 R-CNN Network Model 3.2 Object Detection and Classification Using R-CNN Network 4 Classification of Semantic Relationship Based on KD-Tree 4.1 Triples Describe a Semantic Relationship 4.2 Building a KD-Tree Structure for Semantic Relationship Classification 4.3 Training Weight Vector Process on KD-Tree 4.4 Extraction Semantic Relationship of an Image Using KD-Tree 5 A Model of Object Classification and Semantic Relationship Extraction 5.1 A Proposed Model 5.2 Multi-object Image Retrieval Based on Semantic Relationships and Ontology 5.3 Experiment and Evaluate the Results 6 Conclusion References Preliminary Study on Video Codec Optimization Using VMAF 1 Introduction 2 Related Work 3 Methodology 3.1 Experiment Setup 3.2 Results 4 Conclusion and Future Work References Semantic-Based Image Retrieval Using RS-Tree and Knowledge Graph 1 Introduction 2 Related Works 3 Architecture of the Semantic-Based Image Retrieval System 4 Image Retrieval System 4.1 Description of RS-Tree Structure 4.2 Knowledge Graph Construction 4.3 Scene Graph 4.4 Algorithm of Content-Based Image Retrieval Using RS-Tree 4.5 Creating of SPARQL Query 4.6 Algorithm of Image Semantic Retrieval 5 Experiments and Discussions 5.1 Experimental Environment 5.2 Experimental Evaluation 6 Conclusion References .26em plus .1em minus .1emAn Extension of Reciprocal Logic for Trust Reasoning: A Case Study in PKI 1 Introduction 2 Related Works 2.1 Trust Relationship and Trust Properties 2.2 Reciprocal Logic and Its Extension 3 A New Extension of Reciprocal Logic 4 A Case Study of Trust Reasoning Based on New Extension in PKI 4.1 Scenario 4.2 Formalization 4.3 Trust Reasoning Process 4.4 Discussion 5 Concluding Remarks References Common Graph Representation of Different XBRL Taxonomies 1 Introduction 2 XBRL Format and Taxonomies 3 Graph Representation 4 Practical Examples 5 Searching XBRL Graph Database 6 Conclusions References Natural Language Processing Development of CRF and CTC Based End-To-End Kazakh Speech Recognition System 1 Introduction 2 Literature Review 3 Methodology of E2E Models 3.1 Connectionist Temporal Classification (CTC) 3.2 Conditional Random Fields (CRF) 3.3 Joint Training Models for Kazakh Speech Recognition 4 Experiments and Results 4.1 Data Preparation 4.2 Presetting Models 4.3 Results and Analysis of Experimental Studies 5 Conclusion References A Survey of Abstractive Text Summarization Utilising Pretrained Language Models 1 Introduction 2 Background 2.1 Neural Abstractive Text Summarization 2.2 Elements of Abstractive Text Summarization Systems 3 Pretrained Language Models and Abstractive Text Summarization 4 State-of-the-Art Models for Abstractive Text Summarization 4.1 Fully Abstractive 4.2 Extractive-Abstractive 5 Analysis and Findings 5.1 Issues and Challenges in Finetuning the PLMs 5.2 Performance Improvement Techniques 6 Summary and Conclusions References A Combination of BERT and Transformer for Vietnamese Spelling Correction 1 Introduction 2 Related Works 3 Our Approach 3.1 Introduction to the Vietnamese Language 3.2 Analyzing of Vietnamese Common Spelling Error 3.3 BERT 3.4 Transformers 3.5 Incorporate BERT into Transformers 4 Experimental Evaluation 4.1 Experimental Dataset 4.2 Evaluating Metric 4.3 Model Settings 4.4 Experimental Results and Discussion 5 Conclusion References Enhancing Vietnamese Question Generation with Reinforcement Learning 1 Introduction 2 Related Work 3 Task Definition 4 Methods 4.1 Baseline Models 4.2 Question Generation Enhanced with Reinforcement Learning 5 Experiments and Results 5.1 Datasets 5.2 Evaluation Metrics 5.3 Implementation Details 5.4 Model Comparison 5.5 Experimental Results 5.6 Question Type Prediction Analysis 6 Conclusion and Future Work References A Practical Method for Occupational Skills Detection in Vietnamese Job Listings 1 Introduction 2 Methodology 3 Implementation 3.1 Phrase Mining 3.2 Text Embedding: Universal Methods for Words, Phrases, Sentences, and Paragraphs 3.3 Term Ranking 3.4 Occupational Skill Classification 4 Experiments and Results 4.1 Data Collection 4.2 Evaluation Methods 4.3 Experiments 4.4 Results 5 Conclusions and Future Work References Neural Inverse Text Normalization with Numerical Recognition for Low Resource Scenarios 1 Introduction 2 Literature Review 2.1 Rule-Based Systems 2.2 Neural Network Models 2.3 Hybrid Models 3 Methodology 3.1 General Framework 3.2 Data Creation Process 3.3 Training Model 3.4 Rule-Based Systems 4 Experiment 4.1 Dataset 4.2 Baseline Models and Hyperparameter Configuration 4.3 Evaluation Metrics 4.4 Result Analysis 5 Conclusion References Detecting Spam Reviews on Vietnamese E-Commerce Websites 1 Introduction 2 Related Works 3 The Dataset 3.1 Dataset Creation Process 3.2 Annotation Guidelines 3.3 Inter-annotators Agreement and Discussion 3.4 Dataset Overview 4 Methodologies 4.1 Task Definition 4.2 Word Embedding 4.3 Deep Neural Network Models 5 Empirical Results 5.1 Baseline Results 5.2 Error Analysis 6 Conclusion References v3MFND: A Deep Multi-domain Multimodal Fake News Detection Model for Vietnamese 1 Introduction 2 M2-ReINTEL: Vietnamese Multi-domain Multimodal Dataset 3 v3MFND: A Deep Multi-domain Multimodal Fake News Dectection Model for Vietnamese 3.1 Multimodal Representation Learning 3.2 Multimodal Fusion 3.3 Learning 4 Experimental Setup 4.1 Dataset 4.2 Baselines 4.3 Experimental Settings 4.4 Evaluation Metrics 5 Results 6 Conclusion References Social Networks and Recommender Systems Fast and Accurate Evaluation of Collaborative Filtering Recommendation Algorithms 1 Introduction 2 Related Work 2.1 Collaborative Filtering Algorithms 2.2 Evaluation Methods 3 Proposed Method 4 Experimental Evaluation 4.1 Datasets 4.2 Algorithms 4.3 Metrics 4.4 Results 5 Conclusions References Improvement Graph Convolution Collaborative Filtering with Weighted Addition Input 1 Introduction 2 Related Works 2.1 Collaborative Filtering 2.2 Graph Neural Networks 3 Proposed Method 3.1 Adjacent Graph and Weighted References Matrix 3.2 Embedding Layers 3.3 Propagation Process 3.4 Prediction and Optimization 4 Experiments 4.1 Datasets Description 4.2 Experimental Settings 4.3 Results 5 Conclusion and Future Works 5.1 Conclusion 5.2 Future Works References Combining User Specific and Global News Features for Neural News Recommendation 1 Introduction 2 Related Work 2.1 Conventional Approaches 2.2 Deep Learning-Based Approaches 3 Our Approach 3.1 News Encoder 3.2 User Encoder 3.3 Global News Encoder 3.4 Click Predictor 4 Experiments 4.1 Datasets and Experimental Settings 4.2 Performance Evaluation 4.3 Effectiveness of Global News Encoder 4.4 Ablation Study 5 Conclusion References Polarization in Personalized Recommendations: Balancing Safety and Accuracy 1 Introduction 1.1 Motivation 1.2 Related Works 1.3 Contributions 2 Modeling User Preference Dynamics in RSs 2.1 Modeling Items Consumption per Category as Timeseries 2.2 Hierarchical Agglomerative Clustering of Category Timeseries 2.3 Measurement and Analysis of the uPG Score 3 Post-constrained uPG for Polarization-Safe Top-N Items Recommendation 3.1 Unconstrained MF-Based RS Model and Initial Results 3.2 User-Item uPG Scores Mapping: Key Enabling Factor to Minimize Polarization in RSs 3.3 Solving The Constrained RS Optimization Problem is NP-hard 3.4 Cost-Effective Post-Constrained RS Model: Approach 1 4 Pre-constrained ALS Algorithm and Comparative Empirical Results 4.1 PreALS Algorithm For Mitigated User Polarization in RSs 4.2 Empirical and Comparative Results 5 Conclusion References Social Multi-role Discovering with Hypergraph Embedding for Location-Based Social Networks 1 Introduction 2 Related Work 3 Problem and Approach 3.1 Problem Statement 3.2 Framework Overview 4 Persona Hypergraph Construction 5 Social Multi-role Aware Representation Learning 5.1 Biased Random Walk Based Sampling 5.2 Embedding Learning 6 Empirical Evaluation 6.1 Experimental Setting 6.2 Performance on Friendship Suggestion 6.3 Performance on POI Recommendation 7 Conclusion References Multimedia Application for Analyzing Interdisciplinary Scientific Collaboration 1 Introduction 2 Science Mapping and Inter- and Multidisciplinarity 3 Data and Method of Mapping 4 Analysis Measures 5 Web Application Design 5.1 Communication 5.2 Visual Layouts 6 Summary and Conclusions References CORDIS Partner Matching Algorithm for Recommender Systems 1 Introduction 2 Systematic Literature Review 3 Data and Methodology 4 Results 4.1 Entity Embedding-Based Model 4.2 Keywords Embedding-Based Model 4.3 Final Recommendation List 4.4 Measures of Recommender System Evaluation 5 Concluding Remarks References Author Index