查看文章

The Internet of Things (IoT) is creating a new evolution in the present and future Internet. The idea of IoT is to establish transmission capacities using a ubiquitous, distributed and diverse gadgets network. The rapid growth of the IoT makes the incorporation and connection of several devices a predominant procedure. The increasing numbers of IoT devices and diverse IoT traffic patterns has created the need for traffic classification methods to provide solutions for IoT applications’ issues. Although it has been presented in many papers and surveys, network traffic classification is still undeveloped well in IoT because of the variations in traffic classifications in IoT and NonIoT gadgets. This paper discusses the arising patterns of IoT network traffic classifications and putting them in practical use. It also presents an overview of traditional traffic classification methods, as well as a discussion with a categorization. This paper evaluated the performance metrics such as accuracy, recall, precision and F1 score for these Machine Learning algorithms: Decision Tree (DT), K-Nearest Neighbors (KNN), Naïve Bayes (NB) and Gradient Boosting (GRB) classifiers. The analysis of normal and attack traffic is done by using WEKA software tools and by utilizing the BoT-IoT dataset [1].