Optimization Algorithms with Machine Learning to Improve Security of Internet of Things

Abstract

The IOT network traffic classification is the approach which helps to analyse IOT network traffic. The network traffic analysis can to various network activities. The network traffic analysis process has various steps which include data input, pre-processing, feature extraction, classification and performance analysis. The various machine learning algorithms is proposed in the previous years but those algorithms are unable to achieve high accuracy. The algorithms which are already proposed is unable to extract features from the dataset. To propose algorithm which can extract features from the dataset and achieve high accuracy for the network traffic classification is the motivation this research work. To achieve high accuracy hybrid optimization algorithm is proposed in this paper which is the combination of genetic and PSO algorithm. The hybrid optimization algorithm extract features and later it will be classified using Random Forest. The proposed model is implemented in python and results is achieved in terms of accuracy, precision, recall.