Network Traffic Classification Using Venn Machines Mikhail Dashevskiy

Abstract — Network Traffic Classification is an important prob-lem with applications in such areas as Viruses/Intrusions De-tection System, Antivirus Software, Quality of Service systems. In this article a probabilistic classification system, the Venn Machine, is described and applied to network traffic classification problem. It is shown that our probabilistic prediction system has good performance, and at the same time it gives estimates of the probability that the prediction is correct and the estimates has the property of validity. I