Ensemble Method For Net Traffic Classification Based On Deep Learning

Abstract

With the rapid development of computer network technology, the Internet has covered all aspects of social life. Nowadays, network technology is widely used in various social fields such as economy, military, education, etc. It promotes the rapid development of society and economy, and at the same time brings unprecedented challenges. The security of information transmission and interaction in cyberspace takes network traffic as the carrier, and network traffic contains a large amount of valuable information. How to perceive the current network status through the analysis of network traffic, discover network abnormalities in time is of great significance for maintaining network security. With the rapid development of deep learning in the field of artificial intelligence, researchers have tried to transfer deep learning methods that shine in computer vision processing, natural language recognition and other fields to the field of network flow detection. This paper proposes an ensemble model based on a convolutional neural network, which is integrated on the basis of the CNN model, which reduces the deviation of each basic model and improves the accuracy of the network stream classification results. The main tasks of this paper are as follows: (1) Experiments on the basic model and the integrated model were carried out on the USTC-TFC2016 data set. (2) The experimental results show that the ensemble model reduces the deviation of the basic model and improves the classification accuracy.