Leveraging Random Forest to Detect Botnet Attacks in IoT Environments
Hussien Alrakah, Yagoub Abbker Adam, Mohammed Hassan Osman Abdalraheem, Phiros Mansur, Shaik Mohammed Rizwan, Ibrahim Al–Shourbaji · International Journal of Computational and Experimental Science and Engineering · 2025
Because of their secrecy and capacity to manage vast networks of hacked devices, botnet assaults have grown into a more serious and severe threat to Internet of Things (IoT) devices. The identification of botnet attacks is extremely difficult due to their spread nature and covert activity. IoT devices usually operate with insufficient security safeguards and are vulnerable to these types of assaults. In recent years, machine learning (ML) techniques have shown a lot of promise for identifying and stopping various kinds of cyberattacks. This study accurately detects botnet attacks in Internet of Things environments using a Random Forest (RF)-based approach. The RF model is evaluated on two publicly available datasets designed specifically for botnet discovery. Experimental results show that RF outperforms several other popular models in terms of F1-score, recall, accuracy, and precision. These outcomes show how resilient and effective the RF algorithm is as a practical and reliable method of enhancing IoT device security.