Ensuring security in smart home IoT networks: An ensemble learning-based approach
M. Vinay Kumar Reddy, Amit Lathigara, S.V. Evangeline Sonia · 2025
The rapid expansion of Internet of Things (IoT) gadgets improves modern life but creates security risks. This study emphasizes smart home IoT device security. An ensemble learning-based anomaly detection system uses intrusion datasets to predict and learn new smart home IoT assault patterns. The paper highlights the rise in IoT-related damaging assaults and how anomalies can damage equipment, creating infiltration opportunities. The research shows that the XGBoost ensemble machine learning (ML) model optimizes anomaly detection better than standard methods in a categorical CICIoT2023 dataset. Statistics and ML strengthen smart home IoT systems through parameter assessments and device trustworthiness based on ML model scores. This solution improves IoT device security by validating it with a publicly available smart home dataset and weather conditions.