Searching Strategy Based Cheetah Optimization for Intrusion Detection System of Internet of Things

The Intrusion Detection System (IDS) is a security technology that helps for detect unauthorized threats. Internet of Things (IoT) devices are computerized device that works with internet such as smart home, smart cities and healthcare etc., The IDS detect the unauthorized threats for secure the devices from malicious attacks. To detect the threats or attacks, traditional methods are fails to classify the normal and attacks. To overcome this problem, the Searching Strategy based Cheetah Optimization with Random Forest (SS-CO-RF) is proposed for select the significant features to detect the attacks of IoT devices. The UNSW-NB15 is a data acquisition used for input to the proposed method. The min-max normalization is a pre-processing technique that used to improve the quality of raw data. The SS-CO is a feature selection algorithm used for select the relevant features and improved the performance of model. The RF classifier is utilized to classify as normal and attack. To estimate performance of the model, evaluation metrics such as accuracy, precision, recall and F1-Score of 98.56%, 98.91%, 98.92% and 98.86% on UNSW-NB15 dataset respectively which compared to existing Grasshopper Optimization Algorithm (GOA).