Machine Learning Algorithms for Intrusion Detection in IoT-enabled Smart Homes.

The Internet of Things provides many useful opportunities and makes everyday life easier. At the same time, when interacting with the Internet of Things, a large amount of confidential information is processed, thereby raising the issue of ensuring security. The presented work researches the process of ensuring the security of smart homes by analyzing network traffic to detect malware, which uses machine learning methods Gradient Descent, K-Nearest Neighbor, Recurrent Neural Networks. The research was conducted using the IoT-23 intrusion detection dataset. To evaluate the methods, the metrics accuracy, recall, precision, and F1 were used. As a result, Gradient Descent achieved the highest performance and can be used for intrusion detection. [ABSTRACT FROM AUTHOR]