Optimized Intrusion Detection in IoT Networks Using Machine Learning for Enhanced Security

The rapid growth of IoT devices has transformed modern technology but has also increased the risk of cyberattacks. Resource-constrained IoT devices are vulnerable because they have limited resources and cannot support complex security mechanisms. Machine learning (ML) is a powerful tool that can help detect and prevent various cyber threats. In this study, a supervised ML approach is proposed to address these challenges. A subset of key features is selected using a Genetic Algorithm (GA), and then the selected features are used to train and test a Random Forest (RF) model. Experimental results show that the proposed intrusion detection system (IDS) achieves over 99% accuracy, while precision and recall reach 100% and 99.80%, respectively, using the RT-IoT2022 dataset with 12 distinct attacks. These results demonstrate that the proposed IDS can enhance security for IoT devices while reducing computational requirements, making the approach both effective and efficient for real-world applications. By combining feature selection with the GA and the RF model, the system can detect and mitigate cyber threats more accurately. This study highlights the potential of ML based IDSs to protect IoT networks against attacks and improve overall security.