A survey on performance evaluation of artificial intelligence algorithms for improving IoT security systems

Area of Science:

• Cybersecurity
• Artificial Intelligence
• Internet of Things

Background:

• Internet of Things (IoT) security is a critical and rapidly evolving field.
• Artificial Intelligence (AI), including Machine Learning (ML) and Deep Learning (DL), shows great promise for enhancing IoT security.
• A significant body of research exists, necessitating a comprehensive review and comparison of AI methods for IoT security.

Purpose of the Study:

• To compare the performance of various AI algorithms for intrusion classification and detection in IoT systems.
• To identify the most suitable AI methods and datasets for securing IoT environments.
• To propose a general taxonomy of AI techniques applied to IoT security.

Main Methods:

• A systematic review and comparative analysis of AI algorithms (ML and DL) applied to IoT security.
• Evaluation of AI methods based on accuracy rates for intrusion detection and classification.
• Analysis of research published between 2018 and 2023.

Main Results:

• The study provides a comparative assessment of AI algorithms used in IoT security.
• Identified satisfactory performance of AI techniques in recent research (2018-2023).
• Developed a novel taxonomy of AI techniques for IoT security classification and detection.

Conclusions:

• AI, particularly ML and DL, is a powerful tool for improving IoT security.
• The proposed taxonomy offers a structured overview for researchers and practitioners.
• Further evaluation of AI and IoT datasets is recommended for future work.