Intelligent Systems and Data Science: First International Conference, ISDS 2023, Can Tho, Vietnam, November 11–12, 2023, Proceedings, Part II (Communications in Computer and Information Science)

This two-volume set constitutes the refereed proceedings of the First International Conference on Intelligent Systems and Data Science, ISDS 2023, held in Can Tho, Vietnam, in November 2023. The 35 full papers and 13 short papers presented were thoroughly reviewed and selected from 123 submissions. They are organized in the following topical sections: ​applied intelligent systems and data science for agriculture, aquaculture, and biomedicine; big data, IoT, and cloud computing; deep learning and natural language processing; intelligent systems. Preface Organization Contents – Part II Contents – Part I Applied Intelligent Systems and Data Science for Agriculture, Aquaculture, and Biomedicine Towards Automatic Internal Quality Grading of Mud Crabs: A Preliminary Study on Spectrometric Analysis 1 Introduction 2 Related Works 3 Research Strategy 4 In-vitro Experiments and Preliminary Results 4.1 Sample Preparation and Experimental Setup 4.2 Results and Discussion 5 Conclusion References Deep Learning for Detection and Classification of Nuclear Protein in Breast Cancer Tissue 1 Introduction 2 Background 2.1 Alteration of Nuclear Protein Ki-67 in Breast Cancer Cells 2.2 Deep Network Models 3 Proposed Method 3.1 Training Phase 3.2 Testing Phase 4 Experiments 4.1 Scenarios and Training Parameters 4.2 Training Results 4.3 Testing Results 5 Conclusion References Ensemble Learning with SVM for High-Dimensional Gene Expression Data 1 Introduction 2 Newton Support Vector Machines 3 Newton Support Vector Machines for Gene Expression Datasets 3.1 Newton Support Vector Machines for Classifying Large Number of Dimensions 3.2 Ensemble Learning with Newton Support Vector Machines for Classifying Large Amounts of High Dimensional Datapoints 4 Evaluation 5 Conclusion and Future Works References A New Integrated Medical-Image Processing System with High Clinical Applicability for Effective Dataset Preparation in ML-Based Diagnosis 1 Introduction 2 Process Flow of Dataset Preparation for ML-Based Medical Image Diagnosis 2.1 Medical Institution 2.2 Collaborative Medical Specialists 2.3 Research Institution 3 System Prototyping for Esthetic Outcome Evaluation in Breast Reconstruction Surgery 3.1 Breast Reconstruction Surgery and Esthetic Outcome Evaluation 3.2 Breast Image Acquisition 3.3 Incremental Anonymization and Filtering 3.4 Preprocessing and Feature Extraction 4 Experimental Results 5 Conclusion References An Intelligent Polarimetric System for High-Throughput and High-Sensitivity Quality Assessment of Commercial Semiconductor Wafers 1 Introduction 2 High-Sensitivity and Fast Scanning Infrared Polariscope (HSF-SIRP) 2.1 Functional System Design 2.2 Implementation of Real-Time Controller 3 Experimental Results 3.1 Waveform Analysis for Realtime Performance of HSF-SIRP 3.2 Comparison of System Performance Between SIRP and HSF-SIRP 4 Conclusion References Big Data, IoT, and Cloud Computing Strengthening Information Security Through Zero Trust Architecture: A Case Study in South Korea 1 Introduction 2 Global Cybersecurity Standards 2.1 Paradigm Shift: Zero Trust Architecture 2.2 Limitations of Traditional Security Architectures 2.3 Standardization in Security: Information Security Management Systems 3 Challenges in Existing Frameworks 4 Proposed Zero Trust K-ISMS Framework 4.1 Overview of Korean ISMSs 4.2 ISMS Checklist with Zero Trust 4.3 Use-Case: Decentralized Identifiers 4.4 In-Depth Analysis of Use-Cases and Examples 5 Empirical Analysis in South Korea 5.1 Zero Trust in the Public Sector 5.2 Zero Trust in South Korea: A Comprehensive Approach 6 Conclusion and Future Directions References An AIoT Device for Raising Awareness About Trash Classification at Source 1 Introduction 2 Related Works 2.1 Trash Image Classification on the Trashnet Dataset 2.2 Previous Studies on Smart Trash Cans 3 The Proposed BEG and Its Implementation 3.1 The Proposed Trash Image Classification Model (BEGNet) 3.2 The Proposed IoT-Based Architecture for Prototyping BEG Device 4 Evaluation 4.1 Datasets 4.2 Experimental Setting 4.3 Experimental Results 4.4 Discussion 5 Conclusion References A Big Data Approach for Customer Behavior Analysis in Telecommunication Industry 1 Introduction 2 Apache Spark Framework 2.1 Architecture 2.2 Spark MLlib 3 Proposed Approach 3.1 Florus - A Framework for Handling Large Dataset 3.2 Cluster Settings 3.3 Customer Behavior Analysis 4 Experiment 4.1 User Clustering 4.2 User Classification 4.3 Regression Analysis 5 Conclusion References An Efficient Cryptographic Accelerators for IoT System Based on Elliptic Curve Digital Signature 1 Introduction 2 Background 2.1 Digital Signature Algorithm (DSA) 2.2 ECC and ECDSA 3 Proposed System 3.1 ECDSA Hierarchical Architecture 3.2 Modular Inversion 3.3 Multiplication Modular 3.4 ECDSA Core 3.5 Integration System 4 Experimental Results 5 Conclusions References Improved Gene Expression Classification Through Multi-class Support Vector Machines Feature Selection 1 Introduction 2 Feature Selection Multi-class Support Vector Machines 2.1 Binary Support Vector Machines for Feature Selection 2.2 Multi-class Support Vector Machines for Feature Selection 3 Experimental Results 3.1 Experimental Setup 3.2 Classification Results 4 Conclusion and Future Works References Deep Learning and Natural Language Processing Transliterating Nom Script into Vietnamese National Script Using Multilingual Neural Machine Translation 1 Introduction 2 Background and Related Works 2.1 Overview of the Nom Script 2.2 Multilingual Neural Machine Translation 2.3 Compact Translation Model 2.4 BERT-Fused NMT 2.5 Multi-stage Fine-Tuning 3 The Proposed Model 4 Experiments 4.1 Datasets 4.2 Setup 4.3 Results 4.4 Discussion 4.5 Error Analysis 5 Conclusion and Future Work References FDPS: A YOLO-Based Framework for Fire Detection and Prevention 1 Introduction 2 Literature Review 3 Methodology 3.1 Framework Architecture 3.2 Embedded System 3.3 Custom Dataset 3.4 Fire Detection Model 3.5 Video Processing and Inference 4 Result and Evaluation 4.1 Result 4.2 Evaluation 5 Discussion and Conclusion References Application of Swin Transformer Model to Retrieve and Classify Endoscopic Images 1 Introduction 2 Related Research 3 Proposed Methods of Classification and Feature Extraction 4 Experimental Data and Method 4.1 Experimental Data 4.2 Experimental Method 4.3 Image Retrieval System Evaluation Measurement 5 Results 6 Conclusion References Face Emotion Using Deep Learning 1 Introduction 2 Related Works 3 Preliminaries 3.1 Overview of VGG-16 Model 3.2 Methodologies and Algorithms for Facial Emotion Recognition 4 Experiment 4.1 Evaluating Based on Testing Data Set. 4.2 Comparative Evaluation Method. 5 Conclusions References The Efficiency of Japanese Character Classification by Attention Mechanism 1 Introduction 2 Related Work 2.1 SOTA on the Kuzushiji-MNIST Dataset 2.2 CAM to Grad-CAM 3 Method 3.1 Baseline Single-Label Classification 3.2 Modifying the Architecture of CNN Models 3.3 Attention-Base Classification Head 3.4 Grad-CAM for Model Explainability 4 Experiment 4.1 Implementation 4.2 Compared with SOTA on Kuzushiji-MNIST 4.3 Model Evaluation 5 Discussion 6 Conclusions and Future Work References Using Deep Learning to Build a Chatbot Supporting the Promotion of Speciality Dishes in Mekong Delta 1 Introduction 2 Mekong Delta Specialties Images Classification 2.1 The Mekong Delta Specialties Images Dataset 2.2 Approaches for Classifying Mekong Delta Specialties Images 3 Experimental Results 4 Conclusion References An Approach for Traffic Sign Recognition with Versions of YOLO 1 Introduction 2 Related Work 3 Methods 3.1 Data Collection 3.2 Data Augmentation 3.3 Models for Traffic Sign Recognition 4 Experimental Results 4.1 Environmental Settings 4.2 Results 5 Conclusion References FaceMask Detection Using Transfer Learning 1 Introduction 2 Related Work 3 Method 3.1 The Proposed Workflow 3.2 Dataset 3.3 Data Pre-processing 3.4 Proposed Model 4 Experimental Results 5 Conclusion References Human Intrusion Detection for Security Cameras Using YOLOv8 1 Introduction 2 Related Works 3 Proposed Approach 3.1 System Architecture 3.2 Evaluation Measures 4 Experimental Results 4.1 Data Sets 4.2 Model Training and Testing Results 4.3 System Interface 5 Conclusions References Intelligent Systems Application of 3-gram to English Statements Posted at Q&A Sites to Obtain Factor Scores 1 Introduction 2 Related Works 3 Previous Works 3.1 Obtaining Factors of Statements (Japanese) 3.2 Estimation of Factor Scores (Japanese) 3.3 Obtaining Factors in English 3.4 Obtaining Factors in English 4 Application of 2-gram to Q&A Statements 4.1 Objective 4.2 N-gram 4.3 Analysis Method with Application of 2-gram (English) 5 Application of 3-gram to English Q&A Statements 5.1 Aim 5.2 Analysis Method of 3-gram (English) 5.3 Estimation Result 6 Consideration 6.1 Interpretation on Standardized Partial Regression Coefficient (SPRC) 6.2 Comparison among N-gram and Morphological Analysis (MA) 7 Conclusion Appendix References Spatial-Temporal Information-Based Littering Action Detection in Natural Environment 1 Introduction 2 Related Work 3 Background 4 Proposed Method 5 Experimental Results 5.1 Dataset 5.2 Results 6 Conclusion References A Novel Ensemble K-Nearest Neighbours Classifier with Attribute Bagging 1 Introduction 2 Related Work 3 Methodology 3.1 Proposed Method 3.2 Mutual Information 3.3 Grace 3.4 Asymptotic Analysis 4 Experimental Results 4.1 Dataset 4.2 Experimental Setup 4.3 Models and Hyperparameters 4.4 Evaluation Matrix 4.5 Results 4.6 Comparison with Existing Solutions 4.7 Discussion 5 Conclusion References Legar: A Legal Statute Identification System for Vietnamese Users on Land Law Matters 1 Introduction 2 Related Works 3 Legar System 3.1 Legal-Masked Fine-Tuning for LegaRBERT Generation 3.2 Legal Dictionary 3.3 Legal-Masked Fine-Tuning Strategy 3.4 LegaR Answering Engine 4 Training Processes in Legar System 4.1 Training Process Description 4.2 K-Utility Metric 5 Experimental Results 5.1 Baselines 5.2 Experiments and Ablation Study 6 Conclusion References An Investigation of a Touch-Based Eye-Tracking System with Smart Feedback and Its Influences on Learning - Simulation of Covid-19 Rapid Test System 1 Introduction 2 Literature Review 2.1 Eye-Tracking in Learning with Simulation System 2.2 Eye-Tracking with Touch-Based Manipulation for Procedure Knowledge 2.3 The Effect of Smart Feedback for Eye-Tracking with Touch-Based Learning 3 System Design 3.1 Touch-Based Covid-19 Simulation System 3.2 Eye-Tracking System 4 Methodology 5 Results and Discussions 5.1 Learning Achievements Between Two Groups 5.2 Comparison of Learning Behaviors between Groups 5.3 Comparison Visual Attention Heap Map Between Two Groups for Step 2 and 5 5.4 The Interaction of Eye Movement and Manipulation 6 Conclusion References Improve Medicine Prescribing Performance Using Recommendation Systems 1 Introduction 2 Proposed System 3 Results and Discussion 3.1 Rules Generation 3.2 System Evaluation 4 Conclusion References Author Index