Published Paper Details:

This paper presents a novel traffic classification scheme to improve classification performance when few training data are available. In the proposed scheme, traffic flows aredescribed using the discredited statistical features and flow correlation information is modeled by bag-of-flow (BoF). We solve the BoF-based traffic classification in a classifier combination framework and theoretically analyze the performance benefit. Furthermore, a new BoF-based traffic classification method is proposed to aggregate the naive Bayes (NB) predictions of the correlated flows. We also present an analysis on prediction error sensitivity of the aggregation strategies. Finally, a large number of experiments are carried out on two large-scale real-world trafficdatasets to evaluate the proposed scheme. The experimentalresults show that the proposed scheme can achieve much betterclassification performance than existing state-of-the-art trafficclassification methods. It is useful to tackle a number of network security problems including lawful interception and intrusion detection. In addition, traffic classification also plays an important role in modern network management, such as quality of service (QoS) control. While traditional traffic classification techniques may rely on the port numbers specified by different applications or the signature strings in the payload of IP packets, modern techniques normally utilize host/network behavior analysis or flow level statistical features by taking emerging and encrypted applications into account. Recently, substantial attention has been paid on the application of machine learning techniques to statistical features based traffic classification.