The Internet Traffic Classification an Online SVM Approach

Accurate network traffic classification is fundamental to numerous network activities, from Quality of Service to providing operators with useful forecasts for long- term provisioning. In this paper, we apply Online support vector machine (SVM) technique for internet traffic identification and compare the result with that of previously applied Naive Bayes kernel estimation in AUCKLAND Vi and Entry data sets. Our results show that Online SVM technique is more robust and accurate than Naive Bayes algorithm. The test error can be limited to 5.81% in Entry data sets. For AUCKLAND Vi data sets, the test error can be limited to 14.05% and greatly outperforms Naive Bayes kernel estimation. We also find that Online SVM technique is more efficient than kernel method during classification.