Network Traffic Feature Selection Based on the Evaluation of Every Attribute

Abstract

Traffic classification is making a significant difference in network resource scheduling, safety analysis and future tendency prediction. As the essential step for machine learning based traffic classification, feature subset selection is often used to realize dimension reduction and redundant information decrease. A three-stage hybrid feature subset selection method is proposed to improve the classification performance of hybrid methods at low evaluation consumption. The proposed algorithm disposes features by the rank in the level of block and also evaluates all the remaining features that are verified as useless to take advantage of the interactions among all the features. Our theoretical analysis and experimental observations reveal that the proposed method selects feature subset with impressive classification performance on every index while depleting fewer evaluations