Hybrid ML Approach for Robust Intrusion Detection in IoT Networks
Maqbool Ahmed, Muhammad Imran Ghafoor, Sibghat Ullah Bazai, Sabirov Sardor, Uzair Aslam Bhatti, Temur Eshchanov · 2025
The use of Internet of Things (IoT) devices has attracted more cyberattacks, which calls for the improvement of traditional IDS. The prior research focused on integrating machine learning techniques for addressing emerging threats with the aim to push the boundaries of IDS more adaptable for network security. However, the findings of existing research focused primarily on performance enhancement. This research aims to examine IDS performance with help of ML techniques. This study improved the detection accuracy and lower false alarm rates using a hybrid method by combining K-Means with SVM. The NSL-KDD dataset validates the efficacy of the proposed approach. the results shows notable increases in detection accuracy and lowered false positives, validating the potential of ML in improving network security.