Analysis of Machine Learning Model for Anomaly and Attack Detection in IoT Devices

Today, everything is evolving to be intelligent, whether it is a smart house or a smart industry, related to the usage of smart device, with referring to Internet of Things (IoT) engagement. As IoT becomes more widespread, there is an increase in security breaches linked to the connectivity of susceptible IoT devices. As a result, it is critical to use, intrusion detection algorithms to prevent attacks that take use of IoT security flaws. Traditional intrusion detection systems do not function well withIoT environments due to the restricted capability of IoT and the unique protocols employed. This research work employs a unique strategy for anomalydetection by using machine-learning models such as AdaBoost classifier to find anomalies and attacks in IoT devices with low resources. DS2OS dataset used in the work contains 357,952 rows and 13 columns.The proposed modelpredicts the highest identification and accuracy rate of about 99.56%,as demonstrated by experimental findings on prominent datasets.