Volume 15, Number 4

Wi-Fi Data Analysis based on Machine Learning

  Authors

Nanumura Gedara Umal Anuraga Nanumura, University of South Wales, United Kingdom

  Abstract

This study proposes using machine learning to improve Wi-Fi network security. As Wi-Fi networks spread from industrial to residential areas, the necessity for strong security has risen. The rise of smart networking, especially in the IoT, has created data security and vulnerability issues. A unique method that uses machine learning to detect abnormalities and probable security breaches in Wi-Fi networks addresses these difficulties. We gather, preprocess, and analyse network data to create a complete dataset. This dataset trains machine learning algorithms to identify and classify network anomalies. Using agile methods, data mining, and machine learning algorithms, we created a Wi-Fi network intrusion detection system (WNIDS) that can detect diverse network threats. The proposed WNIDS contains two linked stages with specific machine learning models. These algorithms accurately classify network data as normal or attack specific. Our technology protects against malicious attacks and provides a robust Wi-Fi network for users across domains by incorporating machine learning. Modern network security dangers were fully understood by surveys and data analysis. The WNIDS was implemented and deployed through a structured system development life cycle. This tool eliminates network weaknesses and advances distant enterprises, offering safe and smooth access for consumers globally.

  Keywords

Machine learning, Wi-Fi network security, data security, vulnerability, dataset.