A Model for Emerging Threats and User Behaviour Variability Detection in Smart Homes Using Hybrid Technique

Source abstract

The increasing adoption of smart home technologies has introduced significant security challenges, particularly in detecting intrusions and safeguarding user privacy. Traditional intrusion detection systems (IDS) often struggle with emerging threats and the variability of user behaviours in dynamic environments. This research proposes a hybrid intrusion detection system (HIDS) that integrates machine learning techniques, including Isolation Forest for anomaly detection, K-Means clustering for behavioural analysis, and Long Short-Term Memory (LSTM) networks for predictive modelling. The system is designed to detect sophisticated cyber threats while minimizing false positives through adaptive behaviour modelling. Implemented using Python and Flask, the proposed IDS continuously monitors smart home environments in real-time, ensuring efficient resource utilization while maintaining high detection accuracy. Experimental results demonstrate a significant improvement in intrusion detection performance compared to existing systems, achieving a detection accuracy of 99.2%. This research contributes to the field of smart home security by providing a robust, scalable, and adaptive solution for detecting unauthorized access and emerging cyber threats