Recent Challenges in Intelligent Information and Database Systems: 114th Asian Conference, ACIIDS 2022 Ho Chi Minh City, Vietnam, November 28–30, 2022 Proceedings

Description

This book constitutes the refereed proceedings of the 14th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2022, held in Ho Chi Minh City, Vietnam, in November 2022.   ​This volume contains 60 peer-reviewed papers selected for poster presentation from 406 submissions. Papers included in this volume cover the following topics: 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. Preface Organization Contents G-Fake: Tell Me How It is Shared and I Shall Tell You If It is Fake 1 Introduction 2 Related Work 3 G-Fake 4 Experiments 5 Concluding Remarks References LDA+: An Extended LDA Model for Topic Hierarchy and Discovery 1 Introduction 2 Related Works 3 Methodology 3.1 Text Preprocessing (1) 3.2 Topic Analysis (2) 4 Experiments 4.1 Environment 4.2 Evaluation Metrics 4.3 Experimental Protocol 5 Conclusion References Schema Formalism for Semantic Summary Based on Labeled Graph from Heterogeneous Data 1 Introduction 2 Related Works 2.1 Summarization Approaches 2.2 Discussion and Limitations 3 Contribution 4 A Graph Data Model 4.1 Data Graph Formalism 5 Summarization-Based Content Model 6 Summarization Based Structure Model 7 Experimentation and Results 7.1 Datasets and Experimental Setup 7.2 Evaluation 8 Conclusion and Future Work References Individual Source Camera Identification with Convolutional Neural Networks 1 Introduction 2 Previous and Related Work 3 State-of-the-art Algorithms and Methods for Individual Source Camera Identification 4 Experimental Evaluation 4.1 Experiment I – Results of Classification 4.2 Experiment II – Comparison of Activation Functions 5 Conclusion References Exploring the Effect of Vehicle Appearance and Motion for Natural Language-Based Vehicle Retrieval 1 Introduction 2 Related Work 2.1 Natural Language-Based Video Retrieval 2.2 Data Augmentation Through Back-Translation 3 Proposed Framework 3.1 Overall Framework 3.2 Feature Extraction 3.3 Movement Analysis 4 Experiment and Results 4.1 Dataset and Evaluation Metric 4.2 Experimental Results 5 Conclusions and Future Works References Shapley Additive Explanations for Text Classification and Sentiment Analysis of Internet Movie Database 1 Introduction 2 Related Work 2.1 Sentiment Analysis (SA) and BERT 2.2 Text to Text Analysis by Shapley Additive Explanations (SHAP) 3 Methodology 3.1 Research Workflow 3.2 Internet Movie Database (IMDB) 4 Experiment and Result 4.1 Experiment Results 5 Conclusions References Explaining Predictive Scheduling in Cloud 1 Introduction 2 Related Work 3 Experimental Methodology 3.1 Dataset Background 3.2 Light Gradient Boosting Machine (LightGBM) 3.3 SHapley Additive exPlanations (SHAP) 4 Missing Values Interpolation 4.1 Setup 4.2 Tested Methods 4.3 Comparison 5 Conclusion and Future Work References Exploring Word Embedding for Arabic Sentiment Analysis 1 Introduction 2 Related Work 2.1 Sentiment Analysis for Tunisian Arabic Scripts 3 Word Embedding 4 Proposed Methodology, Results and Discussion 4.1 Feature and Dataset Used 4.2 Approach 4.3 Results 4.4 Discussion 5 Conclusion References Improving Autoencoders Performance for Hyperspectral Unmixing Using Clustering 1 Introduction 1.1 Related Work 2 Method 2.1 Linear Mixing Model 2.2 Autoencoder 2.3 Grid Clustering 2.4 Step-by-Step Pipeline Description 3 Experiments and Results 3.1 Datasets 3.2 Parameters 3.3 Synthetic Data Experiment 3.4 Experiment Scheme 3.5 Results 3.6 Discussion 4 Conclusions References Data-Driven Resilient Supply Management Supported by Demand Forecasting 1 Introduction 2 Related Works 3 Integrating Supply Chain Management with ML 4 Machine Learning Framework 4.1 Evaluated Machine Learning Algorithms 5 Experimental Study 5.1 Data Set 5.2 Results 6 Summary References Machine Learning-Based Recommender System for Tweeting Factory in Industry 5.0 Paradigm 1 Introduction 2 Related Works 3 Problem Description 3.1 Production Plant as an Input-Output Decision-Making Object 3.2 Problem Statement 4 Recommendation Algorithm 5 Architecture of the Recommender System 5.1 Data Collection Module (Acquisition) 5.2 Data Storage (Archiving) Module 5.3 Recommendation and Optimisation Module 5.4 Visualisation Module 5.5 Recommendation Monitoring 6 Experiments 6.1 Applied Methods 6.2 Experiment Methodolody 6.3 Results 7 Final Remarks References Automated Late Fusion of Low Level Descriptors for Feature Extraction and Texture Classification Using Data Augmentation 1 Introduction 2 Related Work 3 Proposed Method 3.1 Overview of the Texture Classification Process 3.2 Program Architecture 3.3 Data Augmentation and Late Fusion Process 4 Experimental Study 4.1 Datasets and Parameter Settings 4.2 Results and Discussion 5 Conclusion References Door-to-Door Sampling Service with Drone 1 Introduction 2 Problem Definition 3 Mathematical Modelling 4 Literature Review 5 Solution Methodology 5.1 Bi-level Genetic Algorithm 5.2 Tabu Search Algorithm 6 Experiments 6.1 Numerical Results and Analyses 6.2 The Benefits of Integrating Drone 7 Conclusion References Error Investigation of Pre-trained BERTology Models on Vietnamese Natural Language Inference 1 Introduction 2 Related Work 3 Dataset 4 Experiments and Results 4.1 Data Preparation 4.2 Experiment Settings 4.3 Experimental Results 5 Result Analysis 5.1 Effects of Annotation Rules 5.2 Effects of Word Overlap 5.3 Effect of Sentence Length 5.4 Hypothesis only Model Analysis 5.5 Error Analysis by Confusion Matrixes 6 Conclusion and Future Work References Supervised Learning Use to Acquire Knowledge from 2D Analytic Geometry Problems 1 Introduction 2 Related Works 3 Solution Analysis 4 Solution Design 4.1 Software Architecture 4.2 Preprocessing 4.3 Training and Testing 4.4 Postprocessing 5 Testing and Validation 6 Conclusions References Forecasting Cryptocurrency Price Fluctuations with Granger Causality Analysis 1 Introduction 2 Background and Literature Review 3 Methodology 3.1 Autoregressive Models 4 Results and Discussion 5 Conclusion References Using Brain-Computer Interface (BCI) and Artificial Intelligence for EEG Signal Analysis 1 Introduction 2 Related Works 3 Methodology 3.1 SEED Dataset 3.2 Mindwave Dataset 3.3 Models Taught and Tested on the Same Group of People 3.4 Models Taught and Tested on Different Groups of People 3.5 Summary 4 Conclusions and Future Work References Predicting Metastasis-Free Survival Using Clinical Data in Non-small Cell Lung Cancer 1 Introduction 2 Materials and Methods 2.1 Patients’ Cohort 2.2 Statistical Methods 3 Results 3.1 Exploratory Data Analysis 3.2 Fifty Percent of Patients Have MFS Below Two years and a Half 3.3 Cox Regression Identified Three Clinical Features that Could Predict MFS 3.4 Cox Regression with Regularization 3.5 The Final Model Predicting Risk of Metastasis Based on Clinical Data 4 Discussion 5 Future Prospects References Impact of Design Decisions on Performance of Embarrassingly Parallel .NET Database Application 1 Introduction 2 Related Works 3 The Research Experiment 3.1 The Evaluation of the Auto-parallelism Implemented in TPL 3.2 Selection of the Optimal Size of Data Chunks 4 The Experiment Results 4.1 Evaluation of the Auto-parallelism Implemented in TPL 4.2 Selection of the Optimal Size of Data Chunks 5 Conclusions and Future Work References Machine Learning to Diagnose Neurodegenerative Multiple Sclerosis Disease 1 Introduction 2 Method 2.1 Understanding the Data 2.2 Data Cleaning 2.3 Preparation of ML Algorithm 3 Results and Discussion 3.1 Exploratory Data Analysis 3.2 ML Results 4 Limitation 5 Conclusion and Future Work References A Novel Integrating Approach Between Graph Neural Network and Complex Representation for Link Prediction in Knowledge Graph 1 Introduction 2 Related Work 2.1 Translational Based Models 2.2 Semantic-Matching Models 3 The Proposed Model 3.1 Graph Aggregation Layers 3.2 The Scoring Function 4 Experiments and Result Analysis 4.1 Datasets and Evaluation Protocol 4.2 Hyperparameters 4.3 Results 5 Conclusion References Expandable-Convolutional-Block Neural Inference with Spatial-Fused Multi-resolution Features and Terse Convolution for IoT-Based Intelligent Automated Systems 1 Introduction 2 Backgrounds 2.1 Convolutional Neural Networks 3 Proposed Model Compression Scheme 3.1 Overall Architecture 3.2 SPP Layer 3.3 TC Layer 4 Case Study 4.1 System Deployment and Experimental Settings 4.2 Expr. 1: Comparisons of Model Accuracy 4.3 Expr. 2: Comparisons of Model Parameters 4.4 Expr. 3: Comparisons of Model Execution Time 5 Conclusions and Future Work References Traffic Management in Smart City 1 Introduction 2 Problem Statement 3 The Proposed Method and Experimental Results 3.1 Functional Scheme of Traffic Control 4 Conclusion References Identifying Non-intuitive Relationships Within Returns Data of a Furniture Online-Shop Using Temporal Data Mining 1 Introduction 2 Dataset, Goal of Analysis, and Related Work 3 Data Mining Framework for Returns 4 Results and Discussion 5 Summary and Outlook References DACE: Did I Catch You at a Good Time? 1 Introduction 2 Related Work 3 Dataset Description 4 Problem Definition 5 Finding Topical Experts 5.1 Expertise Using Type of Participation 5.2 Expertise Using Community Perception 5.3 Expertise Using Temporal Participation 6 Finding Extended Topical Experts 6.1 Finding Related Tags Using Tag Network Analysis 6.2 Extended Topical Expertise 7 Evaluation 8 Conclusion References Analysis of Dynamics of Emergence and Decline of Scientific Ideas Based on Optimistic and Pessimistic Fuzzy Aggregation Norms 1 Introduction 2 Characteristics of Data Distribution 2.1 Fuzzy Logic 2.2 Gestalt Psychology 3 Methods 3.1 Knowledge Dissemination Coefficient 3.2 Dynamic Behavior of Citations 4 Analysis of the Dynamics of Citations of Influential Articles on Scientific Fields 4.1 Knowledge Dissemination Coefficient 4.2 Estimation of Citation Dynamics 5 Simulation of the Model of the Knowledge and Confidence Dissemination 6 Discussion 7 Conclusions References Fusing Deep Learning with Support Vector Machines to Detect COVID-19 in X-Ray Images 1 Introduction 1.1 Related Work 1.2 Contribution 2 Fusing Deep Learning with Support Vector Machines 3 Experimental Setup 3.1 Selected Deep Architectures and Features 3.2 COVID-19 Datasets 4 Experimental Results 5 Conclusions and Future Work References Image-Based Contextual Pill Recognition with Medical Knowledge Graph Assistance 1 Introduction 2 Related Work 3 Proposed Approach 3.1 Overview 3.2 Prescription-Based Medical Knowledge Graph 3.3 Visual Processing Procedure 3.4 Data Fusion 3.5 Loss Functions 4 Performance Evaluation 4.1 Experimental Setup 4.2 Comparison with Baselines 4.3 Ablation Study 5 Conclusion and Future Work References A Legal Information Retrieval System for Statute Law 1 Introduction 2 Related Work 2.1 The COLIEE Competition 2.2 Transformer-Based Pre-trained Models 3 Methodology 3.1 Fine-Tuning a Transformer-Based Model for Task 3 3.2 Determining Specific-Scenario Queries and Training Specific-Scenario Model 3.3 Addressing Long Articles and Queries 3.4 Ensemble the Retrieval Results 4 Experiments 4.1 Dataset 4.2 Experimental Results 5 Conclusion References Exploring Retriever-Reader Approaches in Question-Answering on Scientific Documents 1 Introduction 2 Related Work 2.1 Transformers Approaches for Long Documents 2.2 Retriever-Reader Methods in Open-Domain QA 3 Method 4 Experiments 4.1 Experimental Setup 4.2 Experimental Results 5 Conclusions References P-FCloHUS: A Parallel Approach for Mining Frequent Closed High-Utility Sequences on Multi-core Processors 1 Introduction 2 Preliminaries 3 Mining Frequent Closed High-Utility Sequences in Parallel 3.1 Multi-core Processors 3.2 Partitioned Result Space Structure 3.3 Single Scan Synchronization Strategy 3.4 The Proposed P-FCloHUS Algorithm 4 Experimental Results 4.1 Influence of Input Thresholds on Runtime Performance and Memory Usage 4.2 Effectiveness of the Single Scan Synchronization Strategy 4.3 Scalability 5 Conclusion and Future Work References dMITP-Miner: An Efficient Method for Mining Maximal Inter-transaction Patterns 1 Introduction 2 Basic Concepts and Related Works 3 The Proposed Algorithm (DMITP-Miner) for Mining FMITPs 3.1 dMITP-Miner Algorithm 3.2 Illustrative Example 4 Experimental Results 4.1 Mining Time 4.2 Memory Usage 5 Conclusion and Future Works References ITCareerBot: A Personalized Career Counselling Chatbot 1 Introduction 2 Related Work 2.1 Context-Aware and Personalized Chatbots 2.2 Career Counselling Chatbots 3 Motivation 4 ITCareerBot - A Personalized Counselling Chatbot 4.1 Architecture 4.2 Context-Aware Knowledge Base 4.3 User Message Analysis 4.4 Response Generator - Personalized Counselling Method 4.5 Implementation and Deployment 5 Experiments 5.1 Effectiveness 5.2 Efficiency and Satisfaction 6 Conclusion and Future Work References Scheduling Parallel Data Transfers in Multi-tiered Persistent Storage 1 Introduction 2 Previous Works 3 Basic Concepts 4 Parallel Processing Plans 4.1 An Overview 4.2 Creating Transfer Plans 4.3 Partitioning of Transfer Plans 4.4 Applying Scheduling Rules 4.5 Generation of Parallel Processing Plans 5 Experiments 6 Summary and Conclusions References Fine-Tuning OCR Error Detection and Correction in a Polish Corpus of Scientific Abstracts 1 Introduction 2 Available Approaches 3 Evaluation of the Out-of-the-Box Solutions 3.1 Development and Evaluation Sets 3.2 Evaluation Method and Results 4 Qualitative Error Analysis 4.1 Symspell 4.2 Microsoft Spellchecker 4.3 Speller 4.4 Google Spellchecker 4.5 ED 3 pl 5 Reuse and Recycle 5.1 Pre-processing 5.2 Adding Missing Polish Letters 5.3 Evaluation of the Proposed Solution 6 Conclusions and Future Work References A Multi-label Classification Framework Using the Covering Based Decision Table 1 Introduction 2 Using Covering Based Decision Table in MLC 2.1 Problem of Determining Label Relationship in MLC Using Covering Based Decision Table 2.2 Definition of Covering Based Decision Table 2.3 An Example of Covering Based Decision Table for the MLC 3 The Multi-label Classification Framework Using the Covering Based Decision Table 3.1 The Proposed Algorithm CDTML-KNN for MLC Using Covering Based Decision Table 3.2 The Multi-Label Classification Framework using the Covering Based Decision Table 4 Experimental Results and Evaluation 4.1 The Datasets 4.2 Experimental Scenarios 4.3 Experimental Results and Discussions 5 Conclusions and Future Work References En-SeqGAN: An Efficient Sequence Generation Model for Deceiving URL Classifiers 1 Introduction 2 Related Works 3 Proposed Method 3.1 The Overall Framework 3.2 En-SeqGAN Model 4 Experiments and Results 4.1 Experiment Setup 4.2 Classifier Fooling by Gray-Box Attack 4.3 Classifier Re-training with GAN-Generated URL Strings 5 Conclusion References Towards Communication-Efficient Distributed Background Subtraction 1 Introduction 2 Related Work 3 Communication-Efficient Distributed Training for Background Subtraction 3.1 Communication-Efficient Framework for Distributed Background Subtraction (CEDFrame) 3.2 Communication-Efficient Distributed Motion Segmentation (D-MoSeg) 4 Experiments and Discussion 5 Conclusion References Critical Factors of Customers Loyalty in E-Commerce 1 Introduction 2 Literature Background 3 Method 3.1 Research Instrument and Sampling 3.2 Data Analysis 4 Factors Shaping Customers’ Loyalty in E-commerce 5 Conclusions Appendix 1 References An Efficient Multi-view Facial Expression Classifier Implementing on Edge Device 1 Introduction 2 Related Works 3 Proposed Architecture 3.1 Four-Stage Light Backbone 3.2 Classifier Module 4 Dataset, Augmentation, and Configuration 4.1 Dataset 4.2 Preprocessing and Augmentation 4.3 Training Configuration 5 Experiments and Results 5.1 Ablative Study 5.2 Comparison on Datasets 5.3 Real-Time Application 6 Conclusions References Keyword Extraction from Short Texts with a Text-to-Text Transfer Transformer 1 Keyword Extraction and Generation 2 The Polish Open Science Metadata Corpus 3 Approaches 3.1 T5, PlT5 and PlT5kw 3.2 FastText and ExtremeText 3.3 TermoPL 3.4 KeyBERT 4 Intrinsic Evaluation 5 Transfer to Other Domains 5.1 News Stories 5.2 Customer Support Dialogues 6 Summary and Future Work References Development Kazakh-Turkish Machine Translation on the Base of Complete Set of Endings Model 1 Introduction 2 Related Works 3 Method 3.1 The Steps for Kazakh-Turkish Machine Translation Based on the CSE Model 3.2 Creating a Complete Set of Endings for the Kazakh and Turkish Languages Using the Morphological Model CSE 3.3 Morphological Analysis of Turkish Endings 3.4 Matching the Morphological Tables of the Kazakh and Turkish Endings 3.5 Create a Corresponding List of Kazakh and Turkish Stems 3.6 Create a Matching List of Kazakh and Turkish Stop Words 3.7 Development of an Algorithm for Machine Translation of Kazakh-Turkish Based on the CSE Model 3.8 Creating Software Based on the Algorithm 4 Experiments and Results 5 Conclusion References Gated 3D-CNN for Action Recognition 1 Introduction 2 Related Works 2.1 Two-Stream Networks 2.2 3D CNN Networks 2.3 (2+1)D CNN Networks 2.4 Attention Mechanism 3 Proposed Method 4 Experiments 4.1 Dataset 4.2 Implementation Details 5 Evaluation 6 Conclusion and Future Work References A Minimal Model of Cancer Growth, Metastasis and Treatment 1 Introduction 2 Modeling Metastasis – A Brief Literature Review 3 The Mathematical Model 4 Results and Analysis 4.1 Sample Clinical Results 4.2 The Number of Compartments 4.3 Death Condition 4.4 Taking into Account Negative Therapy Effects 5 Conclusions References Utility Driven Job Selection Problem on Road Networks 1 Introduction 2 Background and Problem Definition 3 Proposed Solution Approaches 3.1 Best First Search Approach 3.2 Solution Approach Based on Nearest Neighbor Search 4 Experimental Details 5 Conclusion and Future Research Directions References The Problem of Detecting Boxers in the Boxing Ring 1 Introduction 2 Detection of Boxers in the Boxing Ring 3 Methodology 4 Experiments 4.1 Experimental Design and Data Set 4.2 Results of Experiments 5 Conclusions References Risk Assessment of Acute Kidney Disease and Chronic Kidney Disease for In-Hospital Patients with Acute Kidney Injury 1 Introduction 2 Literature Review 2.1 Literature Gaps 2.2 Gaps Between Guideline and Real Practice: Opportunity for Innovation 2.3 Biomarkers to Facilitate Risk Stratification in AKD 3 Proposed Method 3.1 Overview 3.2 Training Stage 3.3 Prediction Stage 4 Experiments 4.1 Data Analysis 4.2 Experiments for Effectiveness 5 Limitations 6 Conclusions References Music Emotion Recognition Based on Term Frequency and Pattern Entropy 1 Introduction 2 Related Work 2.1 General Multi-labeling Methods 2.2 Deep Learning Methods 3 Proposed Method 3.1 Core Idea 3.2 Framework 3.3 Offline Modeling 3.4 Online Prediction 4 Experiment 4.1 Experimental Data 4.2 Experimental Results 5 Conclusions References Hybrid Boustrophedon and Partition Tree Group Algorithm for Coverage Path Planning Problem with Energy Constraints*-4pt 1 Introduction 2 Related Work 3 Problem Formulation 4 B-WZone Coverage Algorithm 4.1 Boustrophedon Decomposition for the Work Environment P 4.2 Grouping Nodes of Partition Tree 4.3 Coverage by Boustrophedon Motion 5 Experiment and Result Analysis 5.1 Experimental Setup 5.2 Experimental Results 6 Conclusion and Future Work References Predictive Modelling of Diseases Based on a Network and Machine Learning Approach 1 Introduction 2 Material and Methods 2.1 Architecture Overview 2.2 Dataset Description 2.3 Building Disease Network 2.4 Analysis of Disease Network Exploitation 2.5 Developing Risk Prediction Model 3 Result and Discussions 3.1 Disease Network Exploitation 3.2 Developing Disease Prediction Model 3.3 Discussion 4 Conclusion References DeepDream Algorithm for Data Augmentation in a Neural Network Ensemble Applied to Multiclass Image Classification 1 Introduction 2 Materials 3 Methods 3.1 InceptionV3 Neural Network Architecture 3.2 DeepDream Algorithm 3.3 Ensemble Neural Networks 4 Process of Algorithm Development and Training of Neural Network Model 5 Conclusions References Auto Machine Learning-Based Approach for Source Printer Identification 1 Introduction 2 Data and Features Description 3 Proposed Method 4 Experiments and Discussion 5 Conclusion and Future Work References EfficientRec: An Unlimited User Scale Recommendation System Based on Clustering and User's Interaction Embedding Profile 1 Introduction 2 Related Works 2.1 Graph Neural Network in Recommend System 2.2 Self-supervised Learning 3 Proposal Model 3.1 Architecture 3.2 Interaction Embedding 3.3 Contrastive Training 3.4 Item Selection Pipeline 4 Offline Experiment 4.1 Experiment Setting 4.2 Experiment Results 5 Online Experiments 6 Conclusion A Online Sequencial Split B Item Selection Pseudocode References Agent Based Model of Elementary School Group Learning 1 Introduction 2 Related Works 3 Observations and the Proposed Model 4 Evaluation of the Model 4.1 Dataset: A Real Group of Students 4.2 Analysis of Dataset 4.3 Agent Simulation 5 Conclusions References Clinically-relevant Summarisation of Cataract Surgery Videos Using Deep Learning 1 Introduction 2 Related Work 3 Cataract Surgery Video Dataset Description 4 Deep Learning Model Selection Experiments 5 Clinical Analysis of Predicted Annotations 6 Conclusion References An Empirical Comparison of Steganography and Steganalysis Solutions 1 Introduction 2 Related Work 3 Methodology 4 Comparison of Steganalysis Solutions 5 Results for Hiding Payload of Different Sizes 6 Conclusion References Cost-Oriented Candidate Screening Using Machine Learning Algorithms 1 Introduction 2 Related Work 3 Problem Statement 4 Methodology 5 Evaluation 5.1 Experimental Settings 5.2 Evaluation Methods and Metrics 5.3 Cost Ratios 6 Results and Discussion 7 Conclusions and Future Work References Design of IoT Based Patient Health Monitoring System 1 Introduction 2 Related Work 3 Architecture of IoT Monitoring System 4 Implementation and Results 5 Discussion and Conclusion References Author Index