Separating Operator Hippopotamus Optimization Algorithm with Support Vector Machine for Optimizing Intrusion Detection in Internet of Things

An Intrusion Detection System (IDS) is considered to identify system attacks and distinguish between normal and abnormal activities. Machine Learning (ML) techniques are enabled to IDS which is significant role identifying Intrusions. This research presents the system design of an IDS, minimize false alarm rate and increase accuracy to detect intrusion. However, enhancing volume of network traffic in Internet of Things (IoT) system makes it difficult to detect cyberattacks accurately because of huge number of features which most are redundant led to inefficiencies in detection performance. To overcome these challenges, Separating Operator-Hippopotamus Optimization Algorithm with Support Vector Machine (SOHOA-SVM) is proposed for enhancing attack detection, which eliminates redundant features and improve accuracy level. In preprocessing, numerical data then labeled data used for normalizations. The SOHOA-SVM achieves accuracy of 99.99% and 99.97% for NSL-KDD and TON-IoT when compared to SVM and Grey Wolf Optimization (GWO) and XGBoost + Mutual Information Thresholding.