Internet Traffic Classification Based on Detecting Community Structure in Complex Network
基于复杂网络社团划分的网络流量分类

In recent years, Internet traffic classification using port based or payload-based methods is becoming increasingly difficult with peer-to-peer(P2P) applications using dynamic port numbers,masquerading techniques, and encryption to avoid detection. Because supervised clustering algorithm needs accuracy of training sets and it can not classify unknown apphcation,we introduced complex network's community detecting algorithm,a new unsupervised classify algorithm, which has previously not been used for network traffic classification. We evaluated this algorithm and compared it with the previously used unsupervised K-means and DBSCAN algorithm, using empirical Internet traces. The experiment results show complex network's community detecting algorithm works very well in accuracy and produces better clusters, besides, complex network's community detecting algorithm need not know the number of the traffic application beforehand.