Survey on BotShield Dual-Stage AI System for Detection of IOT Botnet Attacks

The Internet of Things (IoT) and the exponential growth in Internet usage have made cyber threats—especially IoT-based botnet attacks—much more sophisticated and dangerous. Firewalls and antivirus software are examples of traditional cybersecurity solutions that are no longer sufficient to address these changing threats. We examine and suggest BotShield, a dual-stage intrusion detection system (IDS) that uses machine learning to detect and stop IoT botnet activity in real time, as a solution to this problem. The first stage involves anomaly detection, which monitors network traffic for abnormal patterns that could signal early botnet activity. The second stage uses classification algorithms to accurately differentiate between benign and malicious traffic. To improve model performance, our approach uses feature selection, normalization procedures, and strong preprocessing techniques. Experimental validation on standard IoT botnet datasets demonstrates high accuracy, low false-positive rates, and real-time detection capability, making BotShield a scalable and efficient solution for modern IoT network security.