Design of an Intrusion Detection Model for IoT-Enabled Smart Home

Machine learning (ML) provides effective solutions to develop efficient intrusion detection system (IDS) for various environments. In the present paper, a diversified study of ensemble machine algorithms has been carried out propose design an and time-efficient IDS Internet Things (IoT) enabled environment. this data captured from network traffic real-time sensors IoT-enabled smart environment analyzed classify predict types attacks. The performance Logistic Regression, Random Forest, Extreme Gradient Boosting, Light Boosting classifiers have benchmarked using open-source largely imbalanced dataset ‘DS2OS’ that consists ‘normal’ ‘anomalous’ traffic. An model “LGB-IDS” proposed LGBM library ML after validating its superiority over other techniques on basis majority voting. is suitably validated certain metrics such as train test accuracy, time efficiency, error-rate, true-positive rate (TPR), false-negative (FNR). experimental results reveal XGB almost equal but efficiency much better than RF, classifiers. main objective paper with high reduced false alarm rate. show achieves accuracy 99.92% comes be higher prevalent algorithms-based models. threat greater 90% less 100%. Time complexity also very low compared algorithms.