An Improved SVM Method for Internet Traffic Classification
Download as PDF
Siyuan Wu, Dengyin Zhang, Fei Ding And Min Zhang
With the development of network technology, network traffic classification technology plays an increasingly important role in network service and management. In terms of classify traffic flows, Support vector machine (SVM) has shown that more effective than traditional methods. However, classification accuracy will have different performance due to different features. Therefore, we proposed a new method to choose the best combination of features. Moreover, a novel SVM is proposed, which can eliminate the influence of noise on classification accuracy. Experimental results show that the method proposed in this paper has a better and more stable performance.
Support vector machine (SVM), features select,network traffic classification, classification accuracy