Filtering-package combination flow feature selection method based on support vector machine

Abstract

The present invention provides a filtering-package combination flow feature selection method based on a support vector machine. The method comprises an initial filtering feature selection method and asecondary package feature selection method embedded with an improved sequence forward search strategy. The initial filtering feature selection method checks contribution of a certain feature on network flow classification and deletes features lower than a set threshold value [Delta] according to the weight of each feature in an original feature set so as to significantly reduce the computation complexity of subsequent feature subset screening; and the secondary package feature selection method embedded with an improved sequence forward search strategy is based on a support vector machine classifier and is embedded with the improved sequence forward search strategy to perform secondary feature selection to select a combination flow feature subset with high distinguishing ability so as to overcome the problems that combination features are deleted by mistake and there is deviation between a feature evaluation result and a final classification algorithm and significantly improve the network flow classification precision. The method is scientific and reasonable, and can be suitable for various flow classification networks.

Description

technical field [0001] The invention belongs to the technical field of computer network traffic classification, and relates to a filter-encapsulation combined traffic feature selection method based on a support vector machine. Background technique [0002] Network traffic classification data often contains many features. These high-dimensional data with many features will lead to increased time and space complexity in the training process, and even produce a "curse of dimensionality", which will completely invalidate existing algorithms. In addition, a large number of redundant and irrelevant features (noise) in high-dimensional data will lead to a sharp decline in the performance of classification models. Feature selection can remove irrelevant features that do not contribute much to the classification results from the original high-dimensional features. Through feature selection, the "curse of dimensionality" can be avoided, the time and space complexity in the algorithm ...

AI Technical Summary

Problems solved by technology

Filtering feature selection directly uses the information and statistical characteristics of the data to evaluate the features. Therefore, its calculation cost is small, and the feature selection speed is fast. It is suitable for processing high-dimensional data, but there are also certain limitations: 1) Redundancy cannot be completely removed feature

2) Poor combination feature selection ability

3) Since the optimal feature subset is selected directly based on the information and statistical characteristics of the data, independent of the learning algorithm, the classification effect is often not very ideal

But its computational complexity is high, if there are n features can generate up to 2 n Feature subsets, using exhaustive search, compare the classification performance of the data set on each subset, when the number of features n is large, exhaustive 2 n feature subset is very difficult

Images