Empowering IoT Security by Classifying Threat Detection Through Machine Learning

Daily lives have been made much easier with the invasion of IoT devices in many fields such as health care, education, smart homes, agriculture, etc. Nowadays these devices are prone to many cyber attacks that exist currently and require impregnable security. The most common attacks that happen on an IoT device are DoS and DDoS. The methodology considered in this research employs machine learning algorithms and PCA (Principal Component Analysis) to classify the type of DDoS and DoS attacks that occur on IoT devices. The research also includes the implementation of genetic algorithm in evolutionary feature selection to optimize the features used in the classification process. This research work employs six supervised machine-learning algorithms such as K-Nearest Neighbor (KNN), Decision Tree, Naive Bayes, Neural Network, Random Forest, and Boosted Tree, to classify the different types of attacks that may occur in the Internet of Things (IoT) devices and a comparative analysis on these algorithms are performed.