Explainable Artificial Intelligence for Cyber Security: Next Generation Artificial Intelligence (Studies in Computational Intelligence, 1025)

Preface
Acknowledgments
Contents
1 The Past, Present, and Prospective Future of XAI: A Comprehensive Review
	1 Introduction
	2 XAI Versus AI
	3 Evolution of XAI
	4 Current Applications of XAI
		4.1 Healthcare
		4.2 Industry
		4.3 Security
		4.4 Natural Language Processing
	5  Usability, User Acceptance, User Trust and XAI
	6 Challenges and Opportunities
		6.1 Challenges
		6.2 Opportunities
	7 Conclusions
	References
2 An Overview of Explainable Artificial Intelligence for Cyber Security
	1 Introduction
	2 Cyber Security
		2.1 The Components of IDS
		2.2 Intrusion Detection Process
		2.3 Detection Mode
		2.4 Strengths and Limitations of IDS
	3 Artificial Intelligence
	4 Explainable Artificial Intelligent for Cyber Security
		4.1 The Performance Evaluation and Pubic Dataset
	5 Discussion
	6 Addressed Challenges for AI in Cyber Security
	7 Conclusion
	References
3 Artificial Intelligence: Practical and Ethical Challenges
	1 Introduction
	2 Machine Learning
	3 Big Data and Machine Learning
	4 Practical Challenges of Machine Learning
		4.1 Sourcing the Data
		4.2 Data Storage and Processing
	5 Ethical Challenges of Machine Learning
		5.1 Data Privacy and Security
		5.2 Bias and Fairness
		5.3 Acceptance and Accountability
	6 Conclusions
	References
4 Domain Knowledge-Aided Explainable Artificial Intelligence
	1 Problem Domain
	2 Methodology
		2.1 Domain Knowledge
		2.2 Domain Knowledge for Intrusion Detection Systems
		2.3 Domain Knowledge Infusion
	3 Experiments
	4 Results
		4.1 Findings from Explainability Test
		4.2 Findings from Generalizability Test
		4.3 Information Sacrifices
		4.4 Explainability Quantification
	5 Discussions and Future Directions
	References
5 Machine Learning Based IDS for Cyberattack Classification
	1 Introduction
	2 Datasets Used in This Research
		2.1 NSL-KDD-Dataset
		2.2 TONIoT Dataset
	3 Machine Learning Models
		3.1 A. Decision Tree
		3.2 Random Forest
		3.3 K-Nearest Neighbours
		3.4 Multi-Layer Perceptron
		3.5 Adaboost
		3.6 XGBoost
	4 Performance Metrics
	5 Results and Discussion
	6 Conclusion
	Appendix
	References
6 Artificial Intelligence for Cyber Security: Performance Analysis of Network Intrusion Detection
	1 Introduction
	2 Literature Review of Existing Works
		2.1 Review of AI-Based Cyber-Security
		2.2 Review of Cloud-Based Cyber-Security
		2.3 Network Monitoring Tool (NMT)
		2.4 Background of Firewall and Various Types of Attacks
	3 System Model
	4 Simulation Results and Discussion
		4.1 Ping Flood DDoS Attack Output
		4.2 TCP SYN Flood DDoS Attack Output:
	5 Conclusion
	References
7 Leveraging Artificial Intelligence Capabilities for Real-Time Monitoring of Cybersecurity Threats
	1 Introduction
	2 Concepts and Background
		2.1 Explainable AI (XAI)
		2.2 Difference Between AI and XAI
		2.3 Cyber Security Threats (CST)
	3 Common Cybersecurity Threats
		3.1 Ransomware
		3.2 Data Leakage and Data Breach
		3.3 Phishing
		3.4 Social Engineering
		3.5 Hacking
		3.6 Insider Threat
		3.7 Denial of Service (DoS) and Distributed Denial of Service (DDoS) Attack
		3.8 Crypto-Jacking
		3.9 Cyber-Physical Attacks
	4 Common Cyber Security Vulnerabilities
		4.1 Use of Broken Algorithms
		4.2 Security Misconfiguration
		4.3 Missing Authorization
		4.4 Missing Authentication for Critical Function (MACF)
		4.5 Failure to Restrict URL Access
		4.6 Insecure Cryptographic Storage
		4.7 Insecure Direct Object References (IDOR)
		4.8 Unrestricted Upload of Dangerous File Types
	5 Real-Time Monitoring (RTM) Tools
		5.1 Threat Maps Based Monitoring Tools
		5.2 Non-Threat Map-Based Monitoring Tools.
	6 Role of AI in Real-Time Monitoring of Cyber Security Threats
		6.1 Event or Incident Prediction
		6.2 Location Intelligence and Situational Awareness
		6.3 Effective Zero-Day Threat Detection
		6.4 Advanced Endpoint Protection
		6.5 Enhanced Vulnerability Management
		6.6 Effective Security Recommendations
	7 Conclusions and Future Work
	References
8 Network Forensics in the Era of Artificial Intelligence
	1 Introduction
	2 Process Model of Network Forensics
	3 Types of Network Attacks
	4 Tools for Network Forensics
	5 Research Advances in Network Forensics
		5.1 IP Traceback Techniques
		5.2 Intrusion Detection Systems
		5.3 Attack Graph-Based Techniques
		5.4 Honeypots
		5.5 Privacy-Preserving Data Analytics
	6 Research Gaps in Network Forensics
		6.1 High Speed and Large Volume of Network Data
		6.2 Network Data Privacy
		6.3 Artificial Intelligence (AI) in Network Forensics
		6.4 Improving Honeypot Development
		6.5 E-Discovery
	7 Explainable Artificial Intelligence for Network Forensics
	8 Conclusion
	References
9 Obfuscation-Based Mechanisms in Location-Based Privacy Protection
	1 Introduction
	2 Performance Metrics and Their Measurement
	3 Implementation and Performance of the Privacy Methods
		3.1 Dataset Used
		3.2 Data Preparation/Exploration
	4 Applying k-Anonymity Methodology with l-Diversity & t-Closeness
		4.1 Privacy Metric
		4.2 Additional Parameters
		4.3 Utility Metric
		4.4 Overhead Metric
	5 Applying Differential Privacy Using K-Means and Logistic Regression Algorithms
		5.1 Differential Privacy using K-Means
		5.2 Privacy Metric
		5.3 Utility Metric
		5.4 Overhead Metric
	6 Differential Privacy with Logistic Regression
		6.1 Privacy Metric
		6.2 Utility Metric
		6.3 Overhead Metric
	7 Applying Dummy Based Location Anonymisation
		7.1 Privacy Metric
		7.2 Utility Metric
		7.3 Overhead Metric
	8 Results
	9 Conclusion
	References
10 Intelligent Radio Frequency Fingerprinting to Identify Malicious Tags in the Internet of Things
	1 Introduction
	2 Background
		2.1 Working Mechanism and Sensory Tag Evolution
		2.2 RFID Security Literature
	3 Overview
		3.1 System and Attack Model
		3.2 Data Collection Process
		3.3 Dataset Explained
		3.4 Novel Techniques Used for Intelligent RF Fingerprinting
		3.5 Gradient Boosted Machine (LGBM)
	4 Proposed Security Framework and Experiment Result
	5 Experiment Result
		5.1 Intelligent RF Fingerprinting
		5.2 Machine Learning to Identify the Malicious Tag and Relative Position
	6 Discussion
	7 Conclusion
	References
11 Explainable Artificial Intelligence for Smart City Application: A Secure and Trusted Platform
	1 Introduction
	2 Smart Cities: The Better Livings for Dwellers
	3 Security Challenges in Smart City Cyber Domain
	4 Classic Artificial Intelligence Methods for Cyber Security in Smart City
		4.1 Fraud Detection
		4.2 Intrusion Detection
		4.3 Spam Detection
		4.4 Malware Detection
		4.5 Traffic Analysis and Identification
	5 Transition to Explainable Artificial Intelligence (XAI)
		5.1 Transportation System
		5.2 Healthcare
		5.3 Smart Grid
	6 Commercial XAI Platforms
	7 Opportunities and Challenges of XAI
	8 Conclusion
	References
12 Explainable Artificial Intelligence in Sustainable Smart Healthcare
	1 Introduction
	2 Why XAI is Important?
	3 XAI in Healthcare
	4 Applications of Artificial Intelligence in Healthcare
	5 Cybersecurity in Healthcare
	6 Big Data in Healthcare
	7 Recent Attacks on Internet of Health Things
	8 XAI for Trustworthy Internet of Health Things
	9 Methods for XAI Covering Cybersecurity and Healthcare
	10 Critical Analysis
	11 Benchmark Datasets for Healthcare Cyber Security
	12 Future Challenges
	13 Conclusions
	References