IoT Network Anomaly Detection in Smart Homes Environment Using Hybrid Machine Learning Approach

The rapid growth of Internet of Things (IoT) devices in smart home environments has led to increased security vulnerabilities and intrusion risks. These resourceconstrained devices require lightweight and effective anomaly detection systems. This research aims to develop a hybrid anomaly detection framework that combines a one-dimensional Convolutional Neural Network (CNN) for feature extraction with LightGBM for classification. The proposed model is evaluated using two benchmark datasets-TON-IoT and UNSW BoT-IoT-based on performance metrics including accuracy, precision, recall, and F1-score. Experimental results demonstrate that the hybrid model outperforms traditional classifiers in both detection accuracy and computational efficiency. This approach provides a lightweight and scalable solution for real-time anomaly detection in smart home IoT networks. This study thus introduces a novel hybrid approach combining CNN-based feature extraction with LightGBM classification, and is optimized for lightweight intrusion detection in IoT smart home environments.