Intelligence of Things: Technologies and Applications: The First International Conference on Intelligence of Things (ICIT 2022), Hanoi, Vietnam, ... Engineering and Communications Technologies)

This book constitutes the refereed proceedings of the First Conference on Intelligence of Things (ICIT 2022), held in Hanoi, Vietnam, in August 2022. A total of 40 full papers in this book have been rigorously peer-reviewed and selected from over 100 submissions. The papers focused on the intelligence of things (AIoT) studies are organized in the following parts: theoretical intelligence analyses, intelligence services and applications, and intelligence service experiments. This book provides interested students and engineers with comprehensive and cutting-edge studies in the fields. Preface Organization Organizing Committee Honorary Chairs General Chairs Program Chairs Steering Committee Track Chairs Liaison Chairs Finance Chair Local Arrangement Chairs Publication Chairs Registration Chairs Multimedia Chairs Program Committee Contents Theoretical Intelligence Analyses Structural Health Monitoring and IoT: Opportunities and Challenges 1 Introduction 2 Internet of Things (IoT) 3 Structural Health Monitoring (SHM) 3.1 Wired and Wireless SHMs 4 IoT Components in SHM 4.1 Sensors Used in SHM 4.2 Data Storage and Processing Hardware 4.3 Communication Protocols 5 Challenges of SHM and IoT 5.1 Energy Consumption 5.2 Scalability 5.3 Security 5.4 Reliability 6 Conclusion References Semantic Interoperability Issues and Challenges in IoT: A Brief Review 1 Introduction 2 Interoperability in IoT 3 Semantic Interoperability 4 Ontology and Semantic Network 5 Related Work on Semantic Interoperability in IoT 5.1 Finding and Discussion on Related Work 6 Challenges in IoT Due to Lack of Semantic Interoperability 7 Open Challenges and Future Work References An Intelligent Method Based on Dissimilarity Measure Picture Fuzzy and Apply to Supplier Selection 1 Introduction 2 Basic Concepts 2.1 Picture Fuzzy Set 2.2 Dissimilarity Between Two Picture Fuzzy Sets 3 Proposed Picture Fuzzy TOPSIS Method 4 Application to Supplier Selection 5 Conclusion References A New Intelligent Computing Method for Scheduling and Crashing Projects with Fuzzy Activity Completion Times 1 Introduction 2 Linear Programming Approach to Project Scheduling and Crashing 2.1 Linear Programming Approach to Project Scheduling (see [16]) 2.2 Linear Programming Approach to Project Crashing (see [16]) 3 Fuzzy PERT: Finding Critical Activities and Critical Path 3.1 Concepts of Flexible Fuzzy Linear Programming 3.2 Calculating the Project Completion Time with Fuzzy Activity Completion Times 3.3 Finding Critical Activities and Critical Path with Fuzzy Activity Completion Times 4 Fuzzy PERT: Crashing the Total Project Completion Time 4.1 Crashing Fuzzy Activity Completion Time Procedure 4.2 Finding Critical Path with Fuzzy Activity Times Under Crashing 5 Concluding Observations References A Study of Filter-Based Feature Selection in Software Fault Prediction 1 Introduction 2 Related Work 3 Background 4 Methodology 4.1 Feature Selection by Filtering 4.2 Learners 4.3 Experimental Setup 4.4 Datasets 4.5 Performance Metrics 5 Experimental Results 6 Conclusion References Applying Lecturer Expectation and Learning Retention to an Adaptive Learning System 1 Introduction 2 Proposed Approach 2.1 Question Classifier 2.2 Determine the Appropriate Difficulty Level for Learners 2.3 Choose Question to Suggest to Learners 3 Implementation and Evaluation 3.1 Implementation Results 3.2 Preliminary Evaluation 4 Conclusion References An Intelligent Approach for Multi-criterial Decision Making Using Similarity of Intuitionistic Fuzzy Sets 1 Introduction 2 Background 3 Entropy of IFSs Based on Similarity Measures 4 Application of Proposed Measure in Ranking Vietnamese Stocks 4.1 Problem Definitions 4.2 Solution Method 5 Experiment and Results 6 Conclusions References Edge Computing for Wireless Sensor Network 1 Introduction 2 Related Works 3 Methodology 3.1 Edge Devices and Machine Learning Models for Comparison 3.2 Energy Model 4 Result and Discussion 4.1 Edge Devices and Machine Learning Models for Comparison 4.2 Energy Calculation 4.3 Conclusion References Mathematical Expression Detection in Camera Captured Document Images 1 Introduction 2 Related Work 2.1 Dewarping Methods of Camera Captured Images 2.2 Detecting MEs in Document Images 3 Proposed System of ME Detection in Camera Captured Images 3.1 Dewarping Input Camera Captured Images 3.2 ME Detection in Camera Captured Images 4 Experimental Results 4.1 Datasets 4.2 Evaluation Metrics 4.3 Performance Evaluation 4.4 Error Analysis and Discussion 5 Conclusion and Future Work References Intelligence Services and Applications A-DEM: The Adaptive Approximate Approach for the Real Scheduling Problem 1 Introduction 2 Related Works 3 Problem Definition 4 Proposed Algorithm 5 Experiment Results 6 Conclusion References A Framework to Verify the ABAC Policies in Web Applications 1 Introduction 2 Preliminaries 2.1 Attribute-Based Access Control 2.2 Spring Security 2.3 A Case Study 3 ABAC Policy Verification 3.1 Confidentiality of Access Policy 3.2 Integrity of Access Policy 3.3 Availability of Access Policy 4 Related Work 5 Conclusion and Future Work References Aspect-Based Sentiment Analysis with Deep Learning: A Multidomain and Multitask Approach 1 Introduction 2 Related Works 3 Methodology 3.1 Convolutional Neural Network (CNN) 3.2 Independently Long Short-Term Memory (IndyLSTM) 3.3 Bidirectional Independently Long Short-Term Memory (Bi-indyLSTM) 3.4 Attention Mechanism 3.5 Proposed Multitask Multidomain Aspect Based Sentiment Analysis Model 4 Experiments and Results 4.1 Dataset 4.2 Experimental Results 5 Conclusion References Combined Local and Global Features for Action Recognition from Motion Sensors 1 Introduction 2 Related Works 3 Method for Action Recognition from Motion Sensors 3.1 Proposed Framework 3.2 Pre-processing 3.3 Feature Extraction 3.4 Classification 4 Experiment 4.1 Datasets 4.2 Experimental Results 5 Conclusion References Combining Artificial Intelligence with the Internet of Things in Microalgae Cultivation 1 Introduction 2 Microalgae Cultivation Technique 2.1 Source of Microalgae Seed 2.2 Factors Affecting the Growth Cycle of Microalgae 2.3 Cultivation Environment 3 Microalgae Cultivation System and Machine Learning Model 3.1 Microalgae Cultivation System 3.2 Machine Learning Model 4 Experimental Results 5 Conclusions References Computer-Aided Provisional Diagnosis System Using Machine Learning 1 Introduction 2 Related Works 3 Proposed Methodology 3.1 Web Application 3.2 Disease Prediction System 4 Experiments 4.1 Dataset 4.2 Classification Strategies 4.3 Evaluation Metrics 4.4 Results 5 Conclusion References Design of Yaw Controller for a Small Unmanned Helicopter Based on Fuzzy Logic Controller and PID-Based Adaptation 1 Introduction 2 Proposed Method 2.1 Fuzzy Logic Controller 2.2 PID-Based Adaptation 3 Software-in-the-Loop and Simulation Results 4 Conclusions References Discovery the Quality of UML Models Through an Exploration of Dataset in Practice 1 Introduction 2 Evaluation Dataset and Tool Support 2.1 Dataset 2.2 Tool Support 3 Design Smells Evaluation 4 Related Work 5 Conclusion References Evaluation of Feature Extraction Methods for Bee Audio Classification 1 Introduction 2 Methodology 2.1 Feature Extraction 2.2 Machine Learning Methods 3 Experiments 3.1 Data Description 3.2 The Performance of the Feature Extractions Methods 4 Concluding Remarks and Perspectives References Indoor Localization Simulation Based on LoRa Ranging 1 Introduction 2 Related Works 2.1 PET in the Indoor Environment 2.2 OMNeT++ Simulator 2.3 Localization with LoRa Approach 3 Indoor Localization in OMNeT++ with LoRa Ranging 3.1 OMNeT++ Configuration 3.2 Errors Filtering 3.3 Indoor Positioning Mechanism in Simulator 4 Result and Discussion 4.1 Result 4.2 Discussion and Future Work 4.3 Conclusion References Mobile Applications Testing Based on Bigraphs and Dynamic Feature Petri Nets 1 Introduction 2 Context-Aware Mobile Apps 3 Related Work 4 A Test Model for Context-Aware Mobile Apps 4.1 Bigraph 4.2 Dynamic Feature Petri Net Model 4.3 Interaction Among Models 5 Case Study 5.1 Modeling the Environment with a Bigraph 5.2 Modeling the Middleware with a DFPN 5.3 Assessments 6 Conclusion and Future Work References Recommendation System with Artificial Intelligence for Welding Quality Improvement 1 Introduction and Related Work 2 Background of RS with AI 2.1 The Welding Quality Improvement System (WQIS) 2.2 RS Architecture of WQIS 3 Methodology 3.1 Data Acquisition 3.2 SVM with Regression for RS 4 Result 5 Conclusion and Future Work References Sentence-Level Sentiment Analysis Using a CNN Model on Contextualized Word Representations 1 Introduction 2 Related Works 3 Research Problems 3.1 Problem Definition 3.2 Research Questions 4 The Proposed Method 4.1 Creating Sentence Embeddings 4.2 Creating Contextualized Word Representations 4.3 Building Sentence-Level Sentiment Classifier 4.4 Training Model 5 Experiment 5.1 Data Acquisition 5.2 Baseline Methods 5.3 Evaluation Results 6 Conclusion and Future Works References Stock Price Prediction in Vietnam Using Stacked LSTM 1 Introduction 2 Methodology 2.1 Long Short-Term Memory 2.2 Stacked LSTM 3 Experiments and Results 3.1 Datasets 3.2 Parameter Settings and Evaluation Metrics 3.3 Results and Discussion 4 Conclusion References Intelligence Service Experiments A High-Performance FPGA-Based Feature Engineering Architecture for Intrusion Detection System in SDN Networks 1 Introduction 2 Background and Related Work 2.1 Software-Defined Networking 2.2 Network Intrusion Detection Systems 2.3 AutoEncoder 2.4 Related Work 3 Methodology 3.1 Overall Architecture 3.2 Data Plane 3.3 Feature Engineering 4 Experiment Results 4.1 Software Validation 4.2 Resource Utilization 4.3 Performance Experiment 4.4 Hardware Reconstruction Data 5 Conclusion References An Approach to Monitoring Solar Farms in Vietnam Using GEE and Satellite Imagery 1 Introduction 2 Study Area and Datasets 3 Google Earth Engine and Algorithms Used 3.1 Google Earth Engine 3.2 CART and Random Forest 4 Related Work 5 Implementation and Result 5.1 General Flow 5.2 Solar Farms Extraction 5.3 Results 6 Conclusions References An FPGA-Based Solution for Convolution Operation Acceleration 1 Introduction 2 Background and Related Work 2.1 Convolution Operation 2.2 Related Work 3 IP Core Architecture 4 Implementation 4.1 Memory Management 4.2 Computing Core 5 Experimental Results 5.1 Synthesis Results 5.2 Simulation Results 6 Conclusion References An Investigation on Vietnamese Credit Scoring Based on Big Data Platform and Ensemble Learning 1 Introduction 2 Related Work 3 Methodologies 3.1 Big Data Platform 3.2 Ensemble Learning 3.3 Important Features Extraction 4 Experiment 4.1 Data Set and Pre-processing 4.2 Experimental Procedure 4.3 Experiment Setup 4.4 Results and Discussion 5 Conclusion and Future Work References Analysis of the Effects of Stop-word Removal in Hate Speech Detection Problem for Vietnamese Social Network Data 1 Introduction 2 Fundamental of Stop-Word Removal and Preprocessing in Hate Speech Detection for Social Network Data 3 Analysis of Stop-Word Removal in Hate Speech Detection 3.1 Procedure of Analysis 3.2 Methods for Stop-Word Selection 3.3 Models 4 Experiment and Analysis 4.1 Datasets 4.2 Collecting Experimental Results 4.3 Analysis of Experimental Results 4.4 Discussion 5 Conclusion References Context Graph Alignment Using Adversarial Learning for Air Pollution Detection on IoT Sensor Systems 1 Introduction 2 Background 2.1 Homogeneous Graph and Heterogeneous Graph 2.2 Network Alignment 3 Unsupervised Learning for Air Pollution Detection on IoT Systems 3.1 Case Study 3.2 Adversarial Learning Training Process 4 Experiment 5 Conclusion References Deep Convolutional Support Vector Machines for Human Activity Recognition 1 Introduction 2 Methods 2.1 1D Deep Convolutional Neural Network 2.2 SVM Algorithm 2.3 Deep Convolutional Support Vector Machines 3 Experimental Details and Results 3.1 Datasets of Human Activities 3.2 Parameter Evaluation 3.3 Classification Results 4 Comparison with Related Works 5 Conclusions References Distributed Deep Learning for Big Data Time Series Analysis 1 Introduction 2 Fundamental of Big Time Series Analysis and Distributed Deep Learning 2.1 Big Data Processing 2.2 Time Series Analysis 2.3 Distributed Deep Learning 3 Proposed System 3.1 Proposed TCN Model 3.2 Distributed Deep Learning Training 4 Experiment 4.1 Experimental Environment 4.2 Experiment Results 5 Conclusion References Safety Helmet Detection at Construction Sites Using YOLOv5 and YOLOR 1 Introduction 2 Methods 2.1 YOLOv5 2.2 YOLOR 2.3 Dataset 2.4 Metrics 3 Results and Discussion 3.1 The Training Process of the Models 3.2 Performance of the Models 4 Conclusion References Image Segmentation of Concrete Cracks Using SegNet 1 Introduction 2 Methodology 2.1 SegNet Network Architecture 2.2 Dataset 2.3 Training Parameter and Evaluation Metric 3 Results and Discussion 4 Conclusion References Intelligent Scene Text Recognition in Streaming Videos 1 Introduction 2 Dataset 3 Method 3.1 SAST 3.2 EAST 3.3 CRNN 3.4 Rosetta 4 Experiment 5 Error Analysis 5.1 Detection 5.2 Recognition References Mangosteen Fruit Detection Using Improved Faster R-CNN 1 Introduction 2 Related Work 3 Data Collection and Preprocessing 4 Proposed Method 4.1 Faster R-CNN 4.2 Mangosteen Images Classifier 5 Experimental Results 6 Conclusion References Naturalness Improvement of Vietnamese Text-to-Speech System Using Diffusion Probabilistic Modelling and Unsupervised Data Enrichment 1 Introduction 2 Background 3 Related Works 3.1 TTS for Vietnamese Language 3.2 Generative Modelling in TTS 4 Unsupervised Data Enrichment 4.1 Data Description 4.2 Data Collection and Unsupervised Data Enrichment 5 Grad-TTS-Based Synthetic Model Approach 5.1 Inference 5.2 Training 6 Experiments 6.1 Experimental Details 6.2 Results 7 Conclusion References Real-Time Traffic Congestion Forecasting Using Prophet and Spark Streaming 1 Introduction 2 Related Work 3 Proposed Architecture 4 Experiments 4.1 Dataset 4.2 Experiment Procedure and Settings 4.3 Metrics 4.4 Result and Discussion 4.5 How Our Framework Works in Action 5 Conclusion References Real-Time Traffic Flow Prediction Using Big Data Analytics 1 Introduction 2 Related Works 3 Proposed Real-Time Traffic Flow Prediction Architecture 3.1 Proposed System Architecture 3.2 Traffic Flow Prediction Model 3.3 Online Real-Time Traffic Flow Prediction Pipeline 4 Experiment 4.1 Experiment Procedure 4.2 Experiment Result and Discussion 5 Conclusion References SSL-MedImNet: Self-Supervised Pre-training of Deep Neural Network for COVID-19 Diagnosis 1 Introduction 2 Related Work 3 Dataset 3.1 Pre-training Unlabeled Data 3.2 Fine-tuning Labeled Data 4 The Proposed Model: SSL-MedImNet 4.1 Self-Supervised Pre-training Using Momentum Contrast 4.2 Supervised Fine-tuning 5 Experimental Results 5.1 Self-Supervised Pre-training 5.2 Supervised Fine-tuning 6 Conclusion References Vietnamese Sentence Paraphrase Identification Using Sentence-BERT and PhoBERT 1 Introduction 2 Related Works 2.1 SBERT Model Architecture 2.2 PhoBERT - A BERT Model for Vietnamese 3 Model 4 Experiments 4.1 Training Dataset 4.2 Training Process 5 Evaluation 5.1 Evaluation on NLI and STS Tasks 5.2 Evaluation on Paraphrase Identification Task 6 Conclusion References Author Index