南京邮电大学学报（自然科学版）2017，Vol.37Issue(2)：80-85,6.DOI:10.14132/j.cnki.1673-5439.2017.02.013
基于粒子群优化算法的视频流特征选择方法
Video traffic feature selection method based on particle swarm optimization
摘要
Abstract
To solve the problem of high feature dimensionality of network video traffic,a feature selection method based on ReliefF and particle swarm optimization (PSO) is proposed.Firstly,some irrelevant features are removed to achieve the fast dimension reduction by ReliefF algorithm.Then,the optimal feature is searched on the subset by using some better subsets of ReliefF algorithm as initial PSO population and evaluating feature subsets by inconsistency rate.The ReliefF algorithm can reduce the feature space and provide the prior knowledge of the PSO algorithm,thus improving the search efficiency and the classification accuracy of the algorithm.Experiments show that the method can achieve better performance in different data sets than existing methods.关键词
特征选择/ReliefF/粒子群优化算法Key words
feature selection/ReliefF/particle swarm optimization (PSO)分类
信息技术与安全科学引用本文复制引用
冯茂,董育宁..基于粒子群优化算法的视频流特征选择方法[J].南京邮电大学学报（自然科学版）,2017,37(2):80-85,6.基金项目
国家自然科学基金(61271233)和华为HIRP创新资助项目 （61271233）
