Advances and Trends in Artificial Intelligence. Theory and Practices in Artificial Intelligence: 35th International Conference on Industrial, ... (Lecture Notes in Computer Science, 13343)

✍ Scribed by Hamido Fujita (editor), Philippe Fournier-Viger (editor), Moonis Ali (editor), Yinglin Wang (editor)

• Publisher: Springer
• Year: 2022
• Tongue: English
• Leaves: 932
• Edition: 1st ed. 2022
• Category: Library

✦ Synopsis

This book constitutes the thoroughly refereed proceedings of the 35th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2022, held in Kitakyushu, Japan, in July 2022.

The 67 full papers and 11 short papers presented were carefully reviewed and selected from 127 submissions. The IEA/AIE 2022 conference focuses on focuses on applications of applied intelligent systems to solve real-life problems in all areas including business and finance, science, engineering, industry, cyberspace, bioinformatics, automation, robotics, medicine and biomedicine, and human-machine interactions.

✦ Table of Contents

Preface
Organization
Contents
Industrial Applications
Comparative Study of Methods for the Real-Time Detection of Dynamic Bottlenecks in Serial Production Lines
1 Introduction
1.1 On the Dynamic Nature of Bottlenecks
1.2 The Need for Real-Time Bottleneck Detection
2 Related Work on Bottleneck Detection
2.1 Detection Using Bottleneck Walk with Buffer Levels
2.2 Detection Using Active Period Method with Machine States
2.3 Detection Using Interdeparture Time Variance with Process Times
3 Design of the Comparative Study for Bottleneck Detection
4 Detection Results using BNW, APM and ITV
4.1 Bottleneck Detection with Bottleneck Walk
4.2 Bottleneck Detection Using the Active Period Method
4.3 Bottleneck Detection Using Interdeparture Time Variances
5 Comparison
5.1 Comparison of 20%-Bottleneck Results
5.2 Results for Varying Bottleneck Process Times (10% to 100%)
6 Conclusion
References
Ultra-short-Term Load Forecasting Model Based on VMD and TGCN-GRU
1 Introduction
2 Methodology
2.1 Variational Mode Decomposition
2.2 Temporal Graph Convolution Network
2.3 VTGG Model
3 Experiments and Discussions
3.1 Data
3.2 Evaluation Method
3.3 Contrast Experimental Model
3.4 Experimental Environment and Parameter Settings
3.5 Experimental Results
4 Conclusion
References
Learning to Match Product Codes
1 Introduction
2 Related Work
3 Data Wrangling
4 Approximate String Matching
5 Deep Learning
6 System Structure Design
7 Experiments and Results
7.1 Exploratory Data Analysis
7.2 Comparison of Approximate String Matching Methods
7.3 Comparison of Deep Learning Methods
8 Conclusion and Future Work
References
ResUnet: A Fully Convolutional Network for Speech Enhancement in Industrial Robots
1 Instruction
2 Related Work
2.1 U-Net
2.2 ResNet
2.3 Huber Loss Function
3 The Proposed Method
3.1 Overview of the Proposed Method
3.2 Structure of Res-Unet
3.3 Optimization Function
4 Experimental Methods
4.1 Dataset
4.2 Feature Transformation
4.3 Training Schemes
4.4 Evaluation Score
5 Experimental Results
6 Conclusion
References
Surface Defect Detection and Classification Based on Fusing Multiple Computer Vision Techniques
1 Introduction
2 Technical Framework
3 Online Defect Detection
3.1 Defect Detection Based on Conventional CV Technology
3.2 Defect Detection Based on CNN
3.3 Detection Result Fusion
4 Offline Defect Classification
5 Case Study and Experiment
5.1 Overall System Architecture
5.2 Data Acquisition
5.3 Online Defect Detection
6 Conclusion
References
Development of a Multiagent Based Order Picking Simulator for Optimizing Operations in a Logistics Warehouse
1 Introduction
2 Order Picking Simulator
2.1 Setting of Simulator
2.2 Cart Behavior Decision Algorithm
3 Experiments for Simulator Performance Evaluation
3.1 Experimental Setting
3.2 Results
4 Discussion
5 Conclusion
References
Health Informatics
Predicting Infection Area of Dengue Fever for Next Week Through Multiple Factors
1 Introduction
2 Related Work
2.1 Study on the Factor of Dengue Fever Model
3 Research Methodology
3.1 Research Characteristics
3.2 Model Scoring
4 Research Experiment
4.1 Data Collection
4.2 Data Preprocessing
4.3 Model Parameter Adjustment
4.4 Experimental Results and Analysis
4.5 Important Characteristics of the Model
4.6 Adjusted Model Results and Analysis
5 Conclusion and Future Research
References
Hospital Readmission Prediction via Personalized Feature Learning and Embedding: A Novel Deep Learning Framework
1 Introduction
2 Basic Notation and Problem Definition
3 The Proposed Framework
3.1 Personalized Feature Learning and Embedding
3.2 Personalized Prediction
4 Experimental Setup
4.1 Dataset Description
4.2 Data Preprocessing
4.3 Baseline Approaches
4.4 Implementation Details and Evaluation Strategies
5 Results and Discussion
5.1 Performance Evaluation
5.2 Clinical Feature Interdependencies
6 Conclusion
References
Intelligent Medical Interactive Educational System for Cardiovascular Disease
1 Introduction
2 Materials and Methods
2.1 Medical Teaching Materials
2.2 Patient-Orient Healthcare Documents
2.3 System Design
2.4 DAG Structure
2.5 Keyword Statistics Architecture
3 Result and Discussion
3.1 Develop a Patient-Centered Educational Interaction System
3.2 Evaluation of Cardiovascular Health Education Data
4 Future Work
References
Evolutionary Optimization for CNN Compression Using Thoracic X-Ray Image Classification
1 Introduction
2 Related Work
2.1 CNN for Xray Images Classification
2.2 Channel Pruning
3 Proposed Method
3.1 Compression-CNN-XRAY
4 Experiments
4.1 Experiment Configuration and Setup
4.2 Results and Discussion
5 Conclusion
References
An Oriented Attention Model for Infectious Disease Cases Prediction
1 Introduction
2 Related Work
3 Problem Definition
4 The Proposed OAM
4.1 Oriented Attention Unit (OAU)
4.2 Temporal Fusion Layer
5 Experiments
5.1 Settings
5.2 Study on Attention Combinations
5.3 Performance Comparisons
6 Conclusions
References
The Differential Gene Detecting Method for Identifying Leukemia Patients
1 Introduction
2 Proposed Method
3 Experiments and Results
4 Conclusions
References
Epidemic Modeling of the Spatiotemporal Spread of COVID-19 over an Intercity Population Mobility Network
1 Introduction
2 The Proposed Approach
2.1 SEIR Model (Single-Network)
2.2 M-Urb-SEIR (Urban Network Epidemic Framework)
2.3 Addressing the Challenges of a Deterministic Epidemic Model
3 Experimental Settings
3.1 Datasets
3.2 Competitors
3.3 Evaluation Metrics
4 Experimental Results
5 Conclusion
References
Skin Cancer Classification Using Different Backbones of Convolutional Neural Networks
1 Introduction
2 Related Work
3 Dataset
4 Model Configuration
5 Experimental Results
6 Conclusion and Future Work
References
Cardiovascular Disease Detection on X-Ray Images with Transfer Learning
1 Introduction
2 Related Work
3 Proposed Method
3.1 Data Pre-processing
3.2 Proposed Model for Cardiovascular Disease Detection
4 Experiments
4.1 Data Set
4.2 Evaluation Methods and Baselines
4.3 Experimental Results
4.4 Discussion on Experimental Results
5 Conclusion
References
Causal Reasoning Methods in Medical Domain: A Review
1 Introduction
2 Probability-Based Reasoning Methods
2.1 Causal Bayesian Networks
2.2 Causal Graph
2.3 Probability Tree
3 Model-Based Reasoning Methods
3.1 SCM
3.2 RCM
3.3 MSM
4 Regression-Based Reasoning Methods
4.1 Granger Causality Test
5 Balancing-Based Reasoning Methods
5.1 Propensity Score Matching
5.2 Re-weighting
5.3 Confounder Balancing
6 Conclusion and Discussion
References
Optimization
Enhancing a Multi-population Optimisation Approach with a Dynamic Transformation Scheme
1 Introduction
2 Related Work
2.1 The Original AMPO Algorithm
2.2 Other Metaheuristic Algorithms
3 The Enhanced Search Framework
4 The Empirical Evaluation
5 Concluding Remarks
References
A Model Driven Approach to Transform Business Vision-Oriented Decision-Making Requirement into Solution-Oriented Optimization Model
1 Introduction
2 Past Related Studies
2.1 Theorical Foundation of MDE
2.2 Previous Experiences in M2M
3 MDE for Decision-Making Process Design
3.1 Cognitive Process for Decision-Making System
3.2 Cognitive Process-Based Model Driven Architecture
4 PIM to PSM Transformation Applied to TSP
4.1 Specification of Solution-Oriented Mathematical Meta-model (SMM)
4.2 Transformation Process
5 Case Study
6 Conclusion and Research Perspectives
References
A Hybrid Approach Based on Genetic Algorithm with Ranking Aggregation for Feature Selection
1 Introduction
2 Related Work
3 Proposed Approach
3.1 The Filter Based Ranking Aggregation
3.2 The RA-GA Algorithm
4 Empirical Settings
5 Experimental Results
5.1 RQ1: How Does the Proposed Approach Perform Comparing with Some State-of-the-Art Methods?
5.2 RQ2: What is the Impact of the Subset's Size Produced by RA-GA?
6 Conclusion
References
A Novel Type-Based Genetic Algorithm for Extractive Summarization
1 Introduction
2 Our Proposed Type-Based GA for Extractive Summarization
2.1 Chromosome Encoder
2.2 Fitness Function
2.3 The Proposed Type-Based GA
3 Related Works
4 Empirical Settings
4.1 Dataset
4.2 Evaluation Metrics
4.3 Tuning Parameters
5 Results
6 Conclusion
References
Dragonfly Algorithm for Multi-target Search Problem in Swarm Robotic with Dynamic Environment Size
1 Introduction
2 Related Works
3 Methodology
3.1 Simulation Parameters Setup
3.2 Environment Setup
4 Results and Discussion
5 Conclusion
References
Video and Image Processing
Improved Processing of Ultrasound Tongue Videos by Combining ConvLSTM and 3D Convolutional Networks
1 Introduction
2 Convolutional LSTM for SSI
3 Data Acquisition and Preprocessing
4 Experimental Setup
5 Results and Discussion
6 Conclusion
References
Improvement of Text Image Super-Resolution Benefiting Multi-task Learning
1 Introduction
2 Related Work
2.1 Scene Text Recognition (STR)
2.2 Scene Text Image Super Resolution
2.3 Multi-task Learning for Super Resolution
3 Methodology
3.1 Text Reconstruction and Super Resolution Transformer (TRSRT)
3.2 Learning with Three Loss Function
4 Experiment
4.1 Datasets and Experiment Settings
4.2 Experimental Results
5 Conclusions
References
Question Difficulty Estimation with Directional Modality Association in Video Question Answering
1 Introduction
2 Related Work
3 Question Difficulty Estimator for Multi-modality
3.1 Overall Structure
3.2 Video Modality Network
3.3 Text Modality Network
3.4 Directional Modality Association Transformer
4 Experiments
4.1 Data Sets
4.2 Baselines
4.3 Experimental Results
5 Conclusions
References
Natural Language Processing
Improving Neural Machine Translation by Efficiently Incorporating Syntactic Templates
1 Introduction
2 Related Works
3 Proposed Template Integration
3.1 Template Generating Methods
3.2 Template Sides
4 Model Architecture
5 Experimental Results
6 Conclusion
References
Forensic Analysis of Text and Messages in Smartphones by a Unification Rosetta Stone Procedure
1 Introduction
2 Related Work
3 Framework
4 An Algorithm for Machine Translation with Rosetta Language Reference, and Double Step-Translation
4.1 A Case Study: Running the Algorithm
5 Conclusions and Further Work
References
Relation-Level Vector Representation for Relation Extraction and Classification on Specialized Data
1 Introduction
2 Previous Works
2.1 Vector Representations of Relations
2.2 Relation Extraction
3 Methodology
3.1 Specialized Corpus and Expert Annotations
3.2 Annotation Representation Unification
3.3 Word Embeddings Training
3.4 Creation by Deduction of Typed Relation Vector Representations
4 Evaluation
4.1 Dataset Balancing
4.2 Typed Relation Identification/Classification
4.3 Comparison to a State-of-the-Art System
4.4 Experimental Results
5 Conclusion and Perspectives
References
SAKE: A Graph-Based Keyphrase Extraction Method Using Self-attention
1 Introduction
2 Related Works
3 Algorithms
3.1 Attention Score of Words Within a Sentence
3.2 Attention Score of Words Cross Sentences
3.3 Graph-Based Ranking
3.4 Forming Candidate Phrases
4 Experimental Settings
4.1 Datasets
4.2 Baselines
4.3 Evaluation Methods
5 Results
6 Conclusion
References
Synonym Prediction for Vietnamese Occupational Skills
1 Introduction
2 Related Work
3 Vietnamese Skill Synonym Dataset
3.1 Phase 1: Skill Collection
3.2 Phase 2: Wikipedia-Based Synonym Suggestion
3.3 Phase 3: Synonym Set Verification
4 Vietnamese Skill Synonym Prediction
4.1 Problem Statement
4.2 Methodology
5 Experiments and Results
5.1 Dataset Construction
5.2 Experiments
5.3 Results
6 Conclusions and Future Work
References
A Survey of Pretrained Embeddings for Japanese Legal Representation
1 Introduction
2 Experimental Settings
3 Experimental Results
4 Further Analysis
5 Conclusions
References
Machine Reading Comprehension Model for Low-Resource Languages and Experimenting on Vietnamese
1 Introduction
2 Proposed Model
2.1 Baseline Model
2.2 Proposal of UtlTran Model
3 Related Works
4 Experiment Preparation
4.1 Datasets
4.2 Evaluation Metrics
4.3 Experimental Setup
5 Experimental Results
5.1 Word Embedding Models
5.2 Translation Strategies
5.3 MRC Models
6 Conclusions
References
Inducing a Malay Lexicon from an Unlabelled Dataset Using Word Embeddings
1 Introduction
1.1 Background
1.2 Motivation and Contribution
2 Related Work
2.1 Lexicon-Based Classification Methodology
2.2 Automatic Lexicon Generation
2.3 Word Embeddings
3 Proposed Framework and Methodology
3.1 Preprocessing
3.2 Lexicon Induction and Lexicon-Based Classification.
3.3 Supervised Document Sentiment Classifier Training and Classification
4 Experimental Results and Discussion
4.1 Dataset
4.2 Lexicon Induction
4.3 Supervised Document Sentiment Classifier
5 Conclusion
References
Agent and Group-Based Systems
Agent-Based Intermodal Behavior for Urban Toll
1 Introduction
2 Background
3 Simulation Framework
3.1 MATSim and eqasim
3.2 Creation of the Intermodal Alternative (car+pt)
3.3 Considering the Road Toll in the Utility Function of the Car
4 Case Study
4.1 Study Area
4.2 Scenario
4.3 Results
4.4 Discussion
5 Conclusion and Perspectives
References
Entropy Based Approach to Measuring Consensus in Group Decision-Making Problems
1 Introduction
2 GDM Problems and Consensus Measurement
3 An Entropic Consensus Measure: Theil-Based Index
4 Comparative Study
5 Conclusion
References
Adaptation of HMIs According to Users’ Feelings Based on Multi-agent Systems
1 Introduction
2 State of the Art
2.1 Model-Based Approaches
2.2 Approaches Based on Fuzzy Logic
2.3 Artificial Intelligence-Based Adaptation Approaches
2.4 Approaches Based on Machine Learning/Deep Learning and Sentiment Analysis
3 Proposed Approach Based on Multi-agent System and Deep Learning
3.1 Presentation of the Proposed Approach
3.2 Proposed Architecture
3.3 Parameters and Rules of the GUI Adaptation for our Application
4 Implementation of the Proposed System
4.1 The Components of the Proposed System
4.2 Examples of Adaptation
5 Experimental Results and Discussion
5.1 Experimentation
5.2 Evaluation and Comparison of the Realized System with Other State-of-the-Art Approaches Based on Sentiment Analysis
6 Conclusion
References
Pattern Recognition
A Generalized Inverted Dirichlet Predictive Model for Activity Recognition Using Small Training Data
1 Introduction
2 Predictive Model
2.1 Generalized Inverted Dirichlet Mixture Model
2.2 Predictive Distribution of the Mixture Model
3 Experimental Results
3.1 Synthetic Data
3.2 Activity Recognition
4 Conclusion
References
Deepfake Detection Using CNN Trained on Eye Region
1 Introduction
2 Background
3 Related Works
3.1 Convolutional Neural Networks
3.2 Generative Adversarial Networks
4 Materials and Methodology
4.1 Dataset
4.2 Frame Extraction
4.3 Face and Eye Region Extraction
4.4 Construct Convolutional Neural Network
4.5 Compare Test Results
5 Results and Conclusion
References
Face Authentication from Masked Face Images Using Deep Learning on Periocular Biometrics
1 Introduction
2 Related Work
3 Methodology
4 Results
5 Conclusions
References
An Optimization Algorithm for Extractive Multi-document Summarization Based on Association of Sentences
1 Introduction
2 Details of Proposed Approach
2.1 Encoding Scheme
2.2 Fitness Function
2.3 Genetic Operations
3 Pseudo Code of Proposed Approach
4 Experimental Result
5 Conclusions and Future Work
References
A Spatiotemporal Image Fusion Method for Predicting High-Resolution Satellite Images
1 Introduction
2 Related Work
3 Proposed Model
3.1 Preprocessing
3.2 Estimation of Missing Pixels (EMP)
3.3 Filtering for Cross-scale Spatial Mismatch (FCSM)
3.4 Modulation of Temporal Changes (MTC)
4 Experimentation and Results
4.1 Comparison with HISTIF
5 Conclusion and Future Work
References
Security
WHTE: Weighted Hoeffding Tree Ensemble for Network Attack Detection at Fog-IoMT
1 Introduction
2 Methodology
2.1 Datasets
2.2 A Lightweight Network Attack Detection System
3 Results and Discussion
4 Conclusions
References
An Improved Ensemble Deep Learning Model Based on CNN for Malicious Website Detection
1 Introduction
2 Literature Review
3 The Proposed CNN-BiGRU Model
4 Results and Discussion
5 Conclusion and Future Work
References
Intrusion-Based Attack Detection Using Machine Learning Techniques for Connected Autonomous Vehicle
1 Introduction
2 Literature Review
3 Methodology
3.1 Dataset Used
3.2 Data Preprocessing
3.3 Feature Engineering
4 Results and Discussion
5 Conclusion
6 Future Scope
References
Detection of Anti-forensics and Malware Applications in Volatile Memory Acquisition
1 Introduction
2 Literature Review
3 Methodology
3.1 Lab Setup
3.2 Analysis/Implementation
4 Results
5 Conclusions
5.1 Limitations
5.2 Contribution
5.3 Future Work
References
Malware Classification Based on Graph Convolutional Neural Networks and Static Call Graph Features
1 Introduction
2 Related Work
3 Graph Convolutional Networks in Malware Analysis
3.1 Graph Convolutional Networks
3.2 Malware Classification Using GCNs
3.3 Comparison: Node-Level Features vs. Topological Features
4 Experimental Results
4.1 Dataset Used
4.2 Results and Discussion
5 Conclusions and Future Work
References
Modelling and Diagnosis
The Java2CSP Debugging Tool Utilizing Constraint Solving and Model-Based Diagnosis Principles
1 Introduction
2 Basic Definitions
3 Java2CSP tool
4 Initial Evaluation
5 Application Scenarios
6 Conclusions
References
Formal Modelling and Security Analysis of Inter-Operable Systems
1 Introduction
2 Related Work
3 Systems Composition Semantics
3.1 Message Passing
3.2 Remote Procedure Call
4 Security Threats Analysis Using SMC-BIP
4.1 Spoofing
4.2 Tampering
4.3 Repudiation
4.4 Information Disclosure (IDS)
4.5 Denial of Service (DoS)
4.6 Elevation of Privilege
5 Conclusion
References
Social Network Analysis
Content-Context-Based Graph Convolutional Network for Fake News Detection
1 Introduction
2 Related Works
3 Problem Definition and Research Questions
4 Proposed Method
4.1 Content and Context Feature Representations
4.2 Construct the C&C Graph from Content and Context Features
4.3 Build the C&C-GCN to Create Node Representations
4.4 Construct the C&C-GCN-Based Fake New Detection Model
5 Experimental Evaluation
5.1 Dataset and Baselines
5.2 Experimental Setup
5.3 Results and Discussion
6 Conclusions and Future Works
References
Multi-class Sentiment Classification for Customers' Reviews
1 Introduction
2 Related Works
3 Datasets
4 Our Proposed Methods
4.1 Data Processing
4.2 Learning Models
4.3 Ensemble Models
5 Experiments
5.1 Data Preparation
5.2 Settings
5.3 Results
6 Conclusion and Future Works
References
Transportation and Urban Applications
MM-AQI: A Novel Framework to Understand the Associations Between Urban Traffic, Visual Pollution, and Air Pollution
1 Introduction
2 Related Work
3 Proposed Framework
3.1 Problem Statements
3.2 Uncertain Temporal Dataset Generation
3.3 Periodic Frequent-Pattern Mining from Uncertain Temporal Databases
3.4 PM2.5 Estimation
3.5 Future PM2.5 Level Prediction
4 Experimental Results
5 Conclusions and Future Works
References
Two-Stage Traffic Clustering Based on HNSW
1 Introduction
2 Related Work
3 Problem Formulation
4 Method
4.1 GPS Trajectory Pre-processing
4.2 Two-Stage Clustering
5 Experiments
6 Conclusion
References
Explainable Online Lane Change Predictions on a Digital Twin with a Layer Normalized LSTM and Layer-wise Relevance Propagation
1 Introduction
2 Related Work
3 Approach
3.1 Lane Change Predictions by a Layer Normalized LSTM
3.2 Explanations of the Prediction Generated by LRP
3.3 Comprehensible Explanations
3.4 Prototype Architecture
4 Evaluation and Discussion
4.1 Evaluation of the Explanations
4.2 Discussion of the Prototype's GUI
5 Conclusion
References
An Agenda on the Employment of AI Technologies in Port Areas: The TEBETS Project
1 Introduction
2 Previous Works
3 The AI Platform
4 TEBETS
5 Development of the Study
5.1 Simulation Model
5.2 Modeling and Simulation (M&S)
5.3 Optimize
6 The Model
7 Discussion
8 Conclusions
9 Future Projects
References
Modelling and Solving the Green Share-a-Ride Problem
1 Introduction
2 Related Work
2.1 The Share a Ride Problem
2.2 Green Transport with Alternative Fuel Station
3 The Green Share-a-Ride Problem Formulation
3.1 Description
3.2 Notations
3.3 The Mathematical Model
4 Numerical Results
5 Conclusion and Future Directions
References
Machine Learning Techniques to Predict Real Time Thermal Comfort, Preference, Acceptability, and Sensation for Automation of HVAC Temperature
1 Introduction
2 Related Work
3 Methodology
3.1 Data Collection
3.2 The Multiclass-Multioutput Classification Model
4 Results and Discussion
5 Conclusion
References
Neural Networks
Serially Disentangled Learning for Multi-Layered Neural Networks
1 Introduction
1.1 Entanglement and Disentanglement
1.2 Information and Entanglement
1.3 Serially Disentangled Learning
2 Theory and Computational Methods
2.1 Serial Disentanglement
2.2 Structural Information
2.3 De-Hierarchical Disentanglement
3 Results and Discussion
3.1 Experimental Outline
3.2 Structural Information and Cost
3.3 Connection Weights and Potentialities
3.4 Compressed Weights
3.5 Summary of Results
4 Conclusion
References
Detecting Use Case Scenarios in Requirements Artifacts: A Deep Learning Approach
1 Introduction
2 Background
2.1 Deep Learning Methods
2.2 Transfer Learning
3 Related Work
4 Method
4.1 Dataset
4.2 Development of Predictive Models for Classifying Use Case Scenario
4.3 Implementation of the Predictive Models
5 Results
5.1 Prediction Performance
5.2 Discussion
5.3 Threat to the Validity
6 Conclusion and Future Work
References
Hybrid Deep Neural Networks for Industrial Text Scoring
1 Introduction
2 Related Work
3 Experimental Framework
3.1 Data Aggregation and Preprocessing
3.2 Modelling
3.3 Evaluation Metric
4 Results and Discussion
4.1 Chosen Regulatory Practices
4.2 Training
4.3 Ablation Study
4.4 Neural Models
4.5 Hybrid Models
4.6 Visualising Word-Level Attention
5 Conclusion
References
Benchmarking Training Methodologies for Dense Neural Networks
1 Introduction
2 Literature Overview
3 Choice of Non-linear Function and Empirical Data Generation
4 Neural Network Models and Training Methodologies
5 Results and Discussion
6 Conclusion and Future Research
References
Proposing Novel High-Performance Compounds by Nested VAEs Trained Independently on Different Datasets
1 Introduction
2 Related Works: Deep Generative Models for Chemical Design
3 Proposal: MatVAE
3.1 Methodological Overview
3.2 Additional Loss Function Lcorr to Amplify Correlation Between Latent Component and Property
4 Evaluation of the Proposed Loss, Lcorr
5 Conclusion
References
Clustering
Monotonic Constrained Clustering: A First Approach
1 Introduction
2 Background
2.1 Constrained Clustering
2.2 Monotonic Clustering
3 The Proposal: Monotonic Constrained Clustering
4 Experimental Setup and Calibration
4.1 Evaluation Method and Validation of Results
4.2 Calibration
5 Experimental Results
6 Statistical Analysis of Results
7 Conclusion
References
Extractive Text Summarization on Large-scale Dataset Using K-Means Clustering
1 Introduction
2 Related Work
3 Model
3.1 Summary Model
3.2 K-Means
3.3 Distance Metrics
3.4 ROUGE
4 Experiment
4.1 Dataset
4.2 Parameters
4.3 The Evaluating Computer
5 Results
5.1 The Summary
5.2 ROUGE Score
5.3 Time
6 Conclusion
References
Multi-Granular Large Scale Group Decision-Making Method with a New Consensus Measure Based on Clustering of Alternatives in Modifiable Scenarios
1 Introduction
2 Preliminaries
3 A MgLSGDM Method with a New Consensus Measure Based on Clustering of Alternatives in Modifiable Scenarios
3.1 Establishing Initial Parameters and Providing Preferences
3.2 Standardisation of Information
3.3 Cluster Creation
3.4 Consensus Calculation
3.5 Aggregation of the Groups' Preference Relations and Obtaining the Ranking of Alternatives
4 A Case Study in E-government
5 Discussion
6 Conclusions
References
Optimal User Categorization from a Hierarchical Clustering Tree for Recommendation
1 Introduction
2 Clustering Extraction from a Clustering Tree
2.1 Agglomerative Hierarchical Clustering
2.2 Framework for the Optimal Selection of Clusters
3 Proposed Method
3.1 Problem of the Sparsity-Based Quality Measure
3.2 Consideration of the Cluster’s Order
3.3 Variance-Based Quality Measure
3.4 General Recommendation Roadmap
4 Experimental Results
4.1 Datasets and Parameter Settings
4.2 Evaluation Metrics
4.3 Comparison Results
5 Conclusions
References
Classification
A Preliminary Approach for using Metric Learning in Monotonic Classification
1 Introduction
2 Background
2.1 Distance Metric Learning
2.2 Monotonic Classification
3 Algorithm Description
3.1 Preliminary Definitions
3.2 Objective Function and Optimization
4 Experiments
4.1 Experimental Framework
4.2 Metrics and Results
4.3 Analysis of Results
5 Conclusion
References
Deep Learning Architectures Extended from Transfer Learning for Classification of Rice Leaf Diseases
1 Introduction
2 Related Work
3 Method
4 Results
5 Conclusion
References
Height Estimation for Abrasive Grain of Synthetic Diamonds on Microscope Images by Conditional Adversarial Networks
1 Introduction
2 Method
2.1 Overview of pix2pix
2.2 Proposed Method
3 Experiments
3.1 Settings
3.2 Results
4 Conclusion
References
Pattern Mining and Tsetlin Machines
Fast Weighted Sequential Pattern Mining
1 Introduction
2 Related Work
3 Preliminaries
4 Proposed FWSPM Algorithm
5 Experimental Evaluation
6 Conclusion
References
Parallel High Utility Itemset Mining
1 Introduction
2 Problem Definition
3 Methodology
3.1 Overall Framework
3.2 Parallel d2HUP
3.3 Parallel EFIM
4 Experiments
4.1 Experiments Design
4.2 Experiment Results
5 Conclusions
References
Towards Efficient Discovery of Stable Periodic Patterns in Big Columnar Temporal Databases
1 Introduction
2 Related Work
3 Model of Stable Periodic-Frequent Patterns
4 Our Mining Algorithm: SPP-ECLAT
4.1 Mining 1-Stable Periodic-Frequent Patterns
4.2 Finding All Interesting Patterns from SPP-ECLAT
5 Experimental Results
6 Conclusions and Future Work
References
Cyclostationary Random Number Sequences for the Tsetlin Machine
1 Introduction
2 Review of the Tsetlin Machine
3 Proposed Random Number Generator in TM
3.1 Randomization in Tsetlin Machine
3.2 Employing Pre-generated Random Number Sequences for TM
3.3 Generating Cyclostationary Random Number Sequences with LFSRs
4 Experimental Results and Discussion
4.1 The Noisy XOR Dataset with Non-informative Features
4.2 The Binary Iris Dataset
5 Conclusions
References
Logics and Ontologies
Evolution of Prioritized EL Ontologies
1 Introduction
2 The Syntax and Semantics of ELs
3 Prioritized EL ontology
3.1 Syntax and Semantics of Prioritized EL Ontology
3.2 Inducing Prioritized EL Ontology
4 Syntactic Evolution of EL Ontology
4.1 Revision with Inconsistent Input
4.2 Revision with Consistent Input
5 Conclusion
References
A Comparison of Resource Data Framework and Inductive Logic Programing for Ontology Development
1 Introduction
2 An Overview: RDF, Prolog and ILP
2.1 RDF
2.2 Prolog
2.3 Inductive Logic Programing
3 Implementation
3.1 Transport Ontology
3.2 Transport Ontology with ILP
3.3 Analogical Analysis of RDF and ILP
4 Conclusion and Future Work
References
MDNCaching: A Strategy to Generate Quality Negatives for Knowledge Graph Embedding
1 Introduction
2 Related Work
2.1 Negative Sampling
3 MDNCaching
3.1 Matrix Decomposition
3.2 The Proposed Strategy
3.3 Integration of MDNCaching with KGE Framework
4 Experiments
4.1 Experimental Setup
5 Conclusion
References
Robotics, Games and Consumer Applications
Application of a Limit Theorem to the Construction of Japanese Crossword Puzzles
1 Introduction
2 Description of Crossword Puzzle Construction and the Application of the Limit Theorem
3 Results and Discussion
4 Conclusion and Future Research
References
Non Immersive Virtual Laboratory Applied to Robotics Arms
1 Introduction
2 Problem Structure
3 Development
3.1 Mathematical Model
3.2 Control Scheme
3.3 Environment Development
3.4 User Interaction with the Virtual Environment
3.5 System Structure
4 Analysis of Results
4.1 Accuracy Virtual Laboratory
4.2 User Improvement Using the Virtual Laboratory
5 Conclusions
References
An Improved Subject-Independent Stress Detection Model Applied to Consumer-grade Wearable Devices
1 Introduction
2 Related Work
3 Stress Detection Dataset
4 Experiments Description
4.1 Bio-signal Processing and Statistical Feature Extraction of EDA, BVP, and ST
4.2 Stress Detection Model Training Methodology
4.3 Experimental Configuration
5 Results
6 Conclusion
References
WDTourism: A Personalized Tourism Recommendation System Based on Semantic Web
1 Introduction
2 Related Works
3 Overall System Architecture
4 Design and Implementation of Core Functions
4.1 The Construction of Scenic Spots Ontology
4.2 Semantic Modeling of Scenic Spot Relationships
4.3 Spatio-Temporal Associative Query Algorithm
4.4 User Personalization Modeling Based on Semantics
5 System Implementation and Results
6 Conclusion
References
Author Index