A Comparative Analysis of Intrusion Detection in IoT Network Using Machine Learning

Recent innovations and advanced technology encourage users to implement solutions against harmful attacks. This is provided new capabilities of dynamic provisioning, monitoring, and management to reduce the IT barriers. IDS is one of the challenging tasks where attackers always change their tools and techniques. Several techniques have been implemented to secure the IoT network, but a few problems are expanding, and their results are not well defined. According to this study, machine learning techniques have been used to detect and classify the problem into the anomaly and normal from the Network Intrusion Detection dataset. First, the data is preprocessed and make it standardize by standard scaler function. The random forest technique has been used to extract the significant features from the dataset. Furthermore, five different classification technique has been used based on the performance measure and compared. The outcome represents that the Decision tree model accomplished the highest accuracy of 100% among other classifiers.