Detecting Botnet Attacks in IoT Healthcare Systems through IoT Technology

The Internet of Things(IoT) has fundamentally changed how technology can be used, but it has also created new opportunities for cyberattacks. These days, there are more IoT devices linked to the network than ever before. IoT device proliferation has led to the emergence and rapid evolution of dangerous attacks including IoT botnet assaults. The study focuses on how IoT devices’ inadequate security features make them susceptible to attacks. By interfering with IoT healthcare system(IoT-HCS) networks and services, these attacks hinder the move to the IoT environment. Particularly, IoT Botnet attacks have become an important risk to IoT-HCS, with hackers employing them to execute widespread attacks on servers and networks. In this situation, Machine Learning(ML) has shown promise as a method for identifying and thwarting these attacks. The paper suggests several ML approaches including Decision Tree(DT), Extreme Gradient Boosting(XGBoost) and Logistics Regression (LR) to detect Botnet attacks in the IoT-HCS. In order to achieve this, the publicly accessible IoT Botnet attack dataset is used. The aforementioned ML techniques have been compared using the IoT Botnet dataset based on a number of evaluation metrics, including recall, F1-score, accuracy, and precision. According to the results, the accuracy rate for the DT model is 98%, while that of the XGboost and LR models is 95% and 89% respectively.