Machine Learning and Probabilistic Graphical Models for Decision Support Systems (2022) [] [9781032039480]

Cover
Title Page
Copyright Page
Preface
Table of Contents
Acronyms
1. Introduction to Machine Learning and Probabilistic Graphical Models for Decision Support Systems
	1 Scope of the Research Domain
	2 Structure of the Book
	3 Conclusion
	References
2. Decision Support Systems for Healthcare based on Probabilistic Graphical Models: A Survey and Perspective
	1 Introduction
		1.1 Probabilistic Modeling
		1.2 Applications of PGMs
	2 Decision Support Systems in Healthcare
		2.1 Probabilistic Graphical Models
		2.2 Bayesian Networks: Directed Graphical Models
		2.3 Markov Random Fields
		2.4 Deep Neural Networks
		2.5 Neural Networks with Probabilistic Graphical Models
	3 Artificial Intelligence in Healthcare Applications
	4 Healthcare Decision Support Systems based on Probabilistic Graphical Models
	5 Perspectives for Healthcare Decision Support Systems based on Probabilistic Graphical Models
	6 Case Studies
		6.1 Logistic Regression for ECG Classification
		6.2 Variational Autoencoder for ECG Anomaly Detection
	7 Conclusions
	References
3. Decision Support Systems for Anomaly Detection with the Applications in Smart Manufacturing: A Survey and Perspective
	1 Introduction
	2 Decision Support Systems for Smart Manufacturing
	3 Anomaly Detection in Smart Manufacturing
		3.1 Smart Predictive Maintenance
		3.2 Integrated Wearable Technology
		3.3 Production Monitoring
		3.4 Real-time Cybersecurity
	4 Difficulties and Challenges of Anomaly Detection Applications in Smart Manufacturing
	5 Perspectives for Anomaly Detection in Smart Manufacturing
	6 Case Studies
		6.1 Anomaly Detection in Production Monitoring
		6.2 Anomaly Detection in Predictive Maintenance
	7 Concluding Remarks
	References
4. Decision Support System for Complex Systems Risk Assessment with Bayesian Networks
	1 Introduction
	2 Bayesian Technology
	3 BN Model for Event Oriented Risk Management
		3.1 Variables Identification
		3.2 Relationships Identification
		3.3 Usage of the model
		3.4 Illustrative Case Study in Natural Risk Management
	4 BN for Risk Management in Industrial Systems
	5 DBN for Risk Management of Industrial Systems
		5.1 Brief Presentation of DBN
		5.2 Illustrative Case Study
	6 EOOBN for Risk Management
		6.1 Extended Object Oriented Bayesian Network
			6.1.1 Construction of an EOOBN
			6.1.2 Case Study
	7 Conclusion
	References
5. Decision Support System using LSTM with Bayesian Optimization for Predictive Maintenance: Remaining Useful Life Prediction
	1 Introduction
	2 Predictive Maintenance and Remaining Useful Life Prediction
	3 Machine Learning based Decision Support System for Predictive Maintenance
	4 Long Short Term Memory Networks using Bayesian Optimization
		4.1 Long Short Term Memory Networks
		4.2 Bayesian Optimization
	5 Decision Support System for Remaining Useful Life Prediction using LSTM with Bayesian Optimization
	6 A Case Study
	7 Conclusion and Perspectives
	References
6. Decision Support Systems for Textile Manufacturing Process with Machine Learning
	1 Introduction
	2 Relevant Literatures
		2.1 Intelligent Techniques used for Textile Process Modeling
			2.1.1 Artificial Neural Networks
			2.1.2 Fuzzy Logic
			2.1.3 Fuzzy Inference System
			2.1.4 Support Vector Machine
			2.1.5 Gene Expression Programming
		2.2 Decision-making of Textile Manufacturing Process
			2.2.1 Classic Methods
			2.2.2 Meta-heuristic Methods
			2.2.3 Multi-criteria Meta-heuristic Methods
	3 Case Study: Decision-making of Denim Ozonation
		3.1 Problem Formulation
		3.2 Methodology
			3.2.1 ANN Model
			3.2.2 Determining the Criteria Weights using the AHP
			3.2.3 The Markov Decision Process
			3.2.4 The RL Algorithm: Q-learning
		3.3 Case Study
			3.3.1 Results and Discussion
	4 Conclusion
	References
7. Anomaly Detection Enables Cybersecurity with Machine Learning Techniques
	1 Introduction
	2 Cybersecurity of Industrial Systems
		2.1 Cyberattack Detection for Industrial Control Systems
		2.2 Anomaly Detection for Time-series Data
	3 Machine Learning-based Anomaly Detection for Cybersecurity Applications
		3.1 Data Driven Hyperparameter Optimization of One-Class Support Vector Machines for Anomaly Detection in Wireless Sensor Networks
			3.1.1 Anomaly Detection Scheme
			3.1.2 Illustrative Example in WSN Anomaly Detection
		3.2 Real Time Data-Driven Approaches for Credit Card Fraud Detection
			3.2.1 Anomaly Detection Scheme
			3.2.2 Illustrative Example in Credit Card Fraud Detection
		3.3 Nested One-Class Support Vector Machines for Network Anomaly Detection
			3.3.1 Nested OCSVMs and Anomaly Detection Scheme
			3.3.2 Illustrative Example in Network Anomaly Detection
		3.4 A Data-Driven Approach for Network Anomaly Detection and Monitoring Based on Kernel Null Space
			3.4.1 Anomaly Detection Scheme
			3.4.2 Illustrative Example in Network Anomaly Detection
	4 Federated Learning-based Anomaly Detection for Cybersecurity Applications
		4.1 Security System Architecture for IoT Systems
			4.1.1 Design of Edge-Cloud System Architecture
			4.1.2 Data Pre-processing at the Edge
			4.1.3 Detection Mechanism
			4.1.4 Performance Evaluation
			4.1.5 Summary
		4.2 Anomaly Detection in Industrial Control System—Smart Manufacturing
			4.2.1 Federated Learning-based Architecture for Smart Manufacturing
			4.2.2 Anomaly Detection Algorithm using Hybrid VAE-LSTM Model at Edge Devices
			4.2.3 Data Pre-processing
			4.2.4 Detection Performance Evaluation
			4.2.5 Evaluation on Edge Computing Efficiency
			4.2.6 Summary
	5 Difficulties, Challenges, and Perspectives for Machine Learning-b ased Anomaly Detection for Cybersecurity Applications
	6 Conclusion
	References
8. Machine Learning for Compositional Data Analysis in Support of the Decision Making Process
	1 Introduction
	2 Modeling of Compositional Data
	3 Machine Learning for Multivariate Compositional Data
		3.1 Principal Component Analysis
		3.2 Clustering
		3.3 Classification
			3.3.1 Support Vector Machine Classification using Ilr—Transformation
			3.3.2 Support Vector Machine Classification using Dirichlet Feature Embedding Transformation
		3.4 Regression
	4 Anomaly Detection using Support Vector Data Description
		4.1 Support Vector Data Description
		4.2 Anomaly Detection using SVDD with Dirichlet Density Estimation
			4.2.1 Transform CoDa using Dirichlet Density Estimation
			4.2.2 Anomaly Detection using SVDD with Dirichlet Density-transformed Data
			4.2.3 An Example of Anomaly Detection using SVDD
	5 Conclusion
	References
9. Decision Support System with Genetic Algorithm for Economic Statistical Design of Nonparametric Control Chart
	1 Introduction
	2 Background
		2.1 Statistical Process Monitoring with Control Chart
		2.2 Parametric and Nonparametric Control Charts
			2.2.1 The x̄ Chart
			2.2.2 The SN Chart
			2.2.3 The SR Chart
		2.3 Related Works
	3 Economic Statistical Design of SN & SR Control Charts
	4 Experiments
	5 Results Discussion
	6 Conclusions
	References
	Appendix
10. Jamming Detection in Electromagnetic Communication with Machine Learning: A Survey and Perspective
	1 Introduction
	2 Electromagnetic Waves Communication Jamming
		2.1 Susceptibility of the Physical Layer in Presence of a Jamming Signal
		2.2 Smart Jamming
	3 Difficulties and Challenges of Electromagnetic Waves Communication Anomaly Detection
		3.1 Detection on Physical Layers
		3.2 Smart Jamming Detection
		3.3 Transmission and Mobility
		3.4 Transmitter Location
	4 Machine Learning Techniques for Electromagnetic Waves Communication Anomaly Detection
		4.1 Classification Algorithms Specificities
		4.2 ML for Jamming Detection Algorithm for a TETRA Base Station Receiver
		4.3 ML for Jamming Detection in 5G Radio Communication
		4.4 ML for Jamming Detection in IoT Network
	5 A Case Study
		5.1 Preliminary Description of the Measurement Test Site
		5.2 Jamming Signals
		5.3 Device Setting
		5.4 Spectrum Analysis
		5.5 Learning and Result
	6 Conclusion
	References
11. Intellectual Support with Machine Learning for Decision-making in Garment Manufacturing Industry: A Review
	1 Introduction
	2 Problems in Garment Manufacturing
	3 Garment Manufacturing using Machine Learning
	4 Popular Machine Learning Algorithms
	5 Potential Machine Learning Applications in Garment Manufacturing
	6 Case Study
	7 Conclusion
	References
12. Enabling Smart Supply Chain Management with Artificial Intelligence
	1 Introduction
	2 AI for Demand Forecasting
	3 AI for Logistics
	4 AI for Production
	5 AI for Decision Support Systems in SCM
	6 Blockchain Technique for SCM
	7 Case Study
	8 Conclusion
	Reference
Index
About the Editor