Identification of network traffic based on support vector machine

Along with the emergence and development of new applications of network which is represented by P2P network, existing network traffic identification methods do not respond to the environment of network traffic which cannot be identified efficiently and accurately. In order to meet the needs of network traffic, the method of network traffic identification based on support vector machine (SVM) is proposed. Over the use of the public data set and the real-time traffic for a combination of supervised learning, this method constructs a reasonable training set and testing set to experiment. And it fully proves that the method for identification of network traffic has high accuracy, low complexity and high recognition efficiency, and the practical feasibility in real-time traffic identification.