Sliding Block-Based Hybrid Feature Subset Selection in Network Traffic

Traffic classification is currently a significant challenge for network monitoring and management. Feature selection is an effective method to realize dimension reduction and decrease redundant information. To realize accurate traffic classification at lower price of evaluations, a hybrid feature subset selection method is proposed on the base of sliding block, the size of which is flexible according to the classification performance. Furthermore, an incremental strategy of convergence is designed on the base of hybrid feature subset selection methods. The strategy gathers all the features that have been selected. To discover the value of relationship among all the selected features, an extra round of selection is added on the base of the original algorithm. The performances are examined by three groups of experiments. Our theoretical analysis and experimental observations reveal that the proposed method consumes fewer evaluations with similar or even better classification performance at different initialized size of block. Moreover, the incremental strategy of convergence makes a further improvement on the classification accuracy.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.