Machine Learning, Blockchain, and Cyber Security in Smart Environments: Application and Challenges (Chapman & Hall/CRC Cyber-Physical Systems)

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Preface
Editors
Contributors
Introduction
Chapter 1: Intelligent Green Internet of Things:: An Investigation
	1.1 Introduction
	1.2 Green IoT
	1.3 Related Surveys
	1.4 IoT Layered Architecture
		1.4.1 Perception Layer
		1.4.2 Transport Layer
		1.4.3 Processing/Middleware Layer
		1.4.4 Network Layer
		1.4.5 Application Layer
	1.5 Applications
		1.5.1 IoT in Industry
		1.5.2 IoT for the Smart Home
		1.5.3 IoT for Agriculture
		1.5.4 IoT in Healthcare
		1.5.5 IoT in Transport
		1.5.6 IoT in Environment and Safety
		1.5.7 IoT in Energy Applications
		1.5.8 IoT in Education
		1.5.9 IoT in Law Enforcement
		1.5.10 IoT in the Prediction of Natural Disasters
		1.5.11 IoT in Consumer Applications
	1.6 IoT Protocols
	1.7 Limitations and Future Research Directions
	1.8 Issues in Energy Conservation
		1.8.1 Idle Listening
		1.8.2 Collision
		1.8.3 Overhearing
		1.8.4 Reduction of Protocol Overheads
		1.8.5 Traffic Fluctuations
	1.9 Energy Preservation Approaches
		1.9.1 Node Activity Management
		1.9.2 Data Aggregation and Transmission Process
		1.9.3 MAC Protocol
		1.9.4 Security Management
		1.9.5 Routing
	1.10 Conclusion
	References
Chapter 2: The Role of Artificial Intelligence in the Education Sector:: Possibilities and Challenges
	2.1 Introduction
	2.2 Background to the Study
		2.2.1 History of AI
		2.2.2 AI in Education
		2.2.3 Applications of AI
		2.2.4 Visions and Challenges of AIED
		2.2.5 EdTech Start-Ups
		2.2.6 Education during Covid-19
	2.3 Literature Survey
	2.4 Findings and Discussion
	2.5 Conclusion
	2.6 Future Work
	Note
	References
Chapter 3: Multidisciplinary Applications of Machine Learning
	3.1 Introduction
	3.2 Machine Learning: A Prolific Concept to Make Machines Learn
		3.2.1 Machine Learning: Workflow
		3.2.2 Prominent Features of Machine Learning
		3.2.3 A Strong Understanding of Machine Learning
		3.2.4 Machine Learning: A Tool Needed at the Right Time
	3.3 Classifications of Machine Learning
		3.3.1 Supervised Learning
		3.3.2 Unsupervised Learning
		3.3.3 Reinforcement Learning
	3.4 Machine Learning in the Modern Era of Computing
	3.5 Application of Machine Learning and Its Relation to Other Fields
		3.5.1 Applying Machine Learning to Agriculture
			3.5.1.1 Pre-Harvesting
			3.5.1.2 Soil
			3.5.1.3 Seeds
			3.5.1.4 Identification of Pesticides and Diseases
			3.5.1.5 Harvesting
			3.5.1.6 Post-Harvest
			3.5.1.7 Vital Parameters to be Considered for an Effective Agricultural Process
		3.5.2 Application of Machine Learning in a Smart HealthCare System for the Elderly in Pandemic Conditions
			3.5.2.1 Smart Healthcare System Architecture
				3.5.2.1.1 Layer 1
				3.5.2.1.2 Layer 2
				3.5.2.1.3 Layer 3
			3.5.2.2 Proposed Smart Healthcare System
				3.5.2.2.1 Phase I
				3.5.2.2.2 Phase II
	3.6 An Overview of Artificial Intelligence and Deep Learning
		3.6.1 Artificial Intelligence
		3.6.2 Deep Learning
	3.7 Conclusion
	References
Chapter 4: Prediction of Diabetics in the Early Stages Using Machine-Learning Tools and Microsoft Azure AI Services
	4.1 Introduction
		4.1.1 Risk Factors for Diabetes
			4.1.1.1 Type 1
			4.1.1.2 Type 2
			4.1.1.3 Pre-Diabetics
			4.1.1.4 Gestational Diabetes
	4.2 Dataset Collection
	4.3 Tools Used for Prediction
		4.3.1 Orange
		4.3.2 RapidMiner
		4.3.3 Microsoft Azure
	4.4 Data Cleansing
		4.4.1 Normalization
		4.4.2 Missing Data
	4.5 Dataset Visualization
		4.5.1 Bar Plot (Gender)
		4.5.2 Bar Plot (HbA1c)
		4.5.3 3D Scatter Plot (HbA1c)
		4.5.4 Bell Curve
	4.6 KNN Implementation
	4.7 Random Forest Implementation
	4.8 Microsoft Azure Implementation
	4.9 Comparison of RapidMiner and Microsoft Azure
	4.10 Conclusion and Future Scope
	References
Chapter 5: Advanced Agricultural Systems:: Identification, Crop Yields and Recommendations Using Image-Processing Techniques and Machine-Learning Algorithms
	5.1 Introduction
	5.2 Literature Survey
	5.3 Proposed Machine-Learning System
	5.4 Dataset
		5.4.1 Data Pre-Processing
		5.4.2 Train-Test Split
		5.4.3 Creating the Classifier Model Using VGG-19
		5.4.4 Evaluating the Model
	5.5 Confusion Matrix
		5.5.1 VGG-19 Model Confusion Matrix
		5.5.2 Train-Test and Validation Loss
	5.6 Classification Algorithm
		5.6.1 XG-boost
		5.6.2 Decision Tree
		5.6.3 Random Forest Classifier
		5.6.4 Naive Bayes Classifier
		5.6.5 Support Vector Machine
	5.7 Conclusion
	References
Chapter 6: SP-IMLA:: Stroke Prediction Using an Integrated Machine-Learning Approach
	6.1 Introduction
		6.1.1 Traditional Risk Factors for Stroke
		6.1.2 Stroke Prevention
	6.2 Problem Statement
	6.3 Motivation
	6.4 Objectives of the Study
	6.5 Review of Relevant Literature
	6.6 Methodology
	6.7 Technology Used
	6.8 Algorithms/Techniques
		6.8.1 K-Means
		6.8.2 Logistic Regression
	6.9 Conclusion and Future Work
	References
Chapter 7: Multi-Modal Medical Image Fusion Using Laplacian Re-Decomposition
	7.1 Introduction
	7.2 Related Work
	7.3 Proposed Methodology
		7.3.1 The Convolutional Neural Network
		7.3.2 Pyramid Decomposition
		7.3.3 Gaussian Pyramid
		7.3.4 Laplacian Pyramid (LP)
	7.4 Fusion Method
	7.5 Results
	7.6 Conclusion
	References
Chapter 8: Blockchain Technology-Enabled Healthcare IoT to Increase Security and Privacy Using Fog Computing
	8.1 Introduction
	8.2 Blockchain with Healthcare IoT and Ethereum
		8.2.1 Distributed Ledger Technologies and Blockchain
		8.2.2 Limitations of Blockchain
	8.3 Supply-Chain Management in Healthcare: Blockchain Technology
		8.3.1 Supply-Chain Management (SCM)
		8.3.2 Healthcare Supply-Chain Management
		8.3.3 Pharmaceutical Supply-Chain Management
		8.3.4 Blockchain-Based Healthcare Companies
			8.3.4.1 Akiri
			8.3.4.2 BurstIQ
			8.3.4.3 Factom
			8.3.4.4 MedicalChain
			8.3.4.5 ProCredEx
	8.4 Genomic Data
		8.4.1 Empowering Genomic Blockchain Technology
		8.4.2 Challenges of Genomics Big-Data Platforms
		8.4.3 Genomic Data Platform Architecture
		8.4.4 Case Study of Genomic Data Sharing in LifeCODE.ai
		8.4.5 Genomic Blockchain Technology
			8.4.5.1 Encrypgen
			8.4.5.2 Health Nexus
			8.4.5.3 Nebula
			8.4.5.4 Opal/Enigma
			8.4.5.5 Shivom
			8.4.5.6 Zenome.io
		8.4.6 Blockchain-Based Clinical Data Sharing FHIRChain
	8.5 Conclusion
	Notes
	References
Chapter 9: Blockchain in Healthcare, Supply-Chain Management, and Government Policies
	9.1 Introduction
	9.2 Blockchain in Healthcare
		9.2.1 Electronic Medical Record Management
		9.2.2 Remote Patient Monitoring
		9.2.3 Medical Supply Chain Management
		9.2.4 Medical Insurance Claims Management
			9.2.4.1 Case Study
		9.2.5 Healthcare Data Protection Management
	9.3 Blockchain in Supply Chain Management
	9.4 Blockchain in Government Policies
	9.5 Conclusion
	References
Chapter 10: Electricity and Hardware Resource Consumption in Cryptocurrency Mining
	10.1 Introduction
		10.1.1 Bitcoin
		10.1.2 How Bitcoin Works
		10.1.3 Organization of the Chapter
	10.2 Literature Survey
	10.3 Methodology
	10.4 Discussion
	10.5 Advantages and Disadvantages of Mining Cryptocurrency
		10.5.1 Advantages of Mining Cryptocurrency
		10.5.2 Disadvantages of Cryptocurrency Mining
	10.6 Conclusion
	10.7 Future Scope
	References
Chapter 11: Cryptographic Hash Functions and Attack Complexity Analysis
	11.1 Introduction
	11.2 Brief Literature Review
	11.3 Comparison of MD5 and SHA1 Based on Collision-Resistant Property
	11.4 Analysis of Dictionary Attack
	11.5 Observations
	11.6 Password Storage Concepts: Salting
	11.7 Statement of the Problem
	11.8 Objectives of the Study
	11.9 Results and Discussion
	11.10 Analyzing the Complexity of Brute-Force Attacks
	11.11 Conclusion and Future Work
	References
Chapter 12: Mixed Deep Learning and Statistical Approach to Network Anomaly Detection
	12.1 Introduction
	12.2 Network Anomaly Detection
	12.3 Traditional Approaches to Network Anomaly Detection
	12.4 Deep Learning Model Flow
	12.5 Preparation of Dataset by Tapping Network Traffic
	12.6 Analysis of Tapped Network Packets Using Wireshark
	12.7 Feature Extraction from a .pcap File
	12.8 Feature Selection
	12.9 Extracting Features from Raw .pcap File
	12.10 Statistical Analysis of Data Set
	12.11 Statistical Anomaly Detection Using Joint Probability Approach
	12.12 Multilayer Perceptron Model
	12.13 Architecture of Multilayer Perceptron
	12.14 Binary Classification of Labels Using PyTorch
	12.15 Architecture of the Multilayer Perceptron Model for Anomaly Classification
	12.16 Improving Model Accuracy Using Statistical Prediction of Anomalies
	12.17 Training the Deep Learning Model Using PyTorch
	12.18 Conclusion and Results
		12.18.1 The Architecture of the Anomaly Detection Model
	12.19 Future Work and Research Propagation
	References
Chapter 13: Intrusion Detection System Using Deep Learning Asymmetric Autoencoder (DLAA)
	13.1 Introduction
	13.2 Literature Survey
	13.3 Methodology
		13.3.1 Autoencoder
		13.3.2 Convolutional Autoencoder
	13.4 Proposed Method
		13.4.1 Asymmetric Convolutional Self-Encoder
		13.4.2 Deep Learning Asymmetric Autoencoder (DLAA)
	13.5 Model Complexity and Timeliness
	13.6 Experiment
	13.7 Experimental Data
	13.8 Data Preprocessing
		13.8.1 Numerical Features
		13.8.2 Normalization
	13.9 Experimental Environment and Parameters
	13.10 Evolutional Index
	13.11 Simulation Experiments and Result Analysis
	13.12 Conclusion and Future Work
	References
Index