基于聚类和二元蚂蚁系统的高维数据特征选择算法  

FEATURE SELECTION ALGORITHM OF HIGH DIMENSIONAL DATA BASED ON CLUSTERING AND BINARY ANT SYSTEM

作　　者：周金容[1] 罗建[2] Zhou Jinrong;Luo Jian(Department of Electronic Information Engineering,Nanchong Vocational and Technical College,Nanchong 637131,Sichuan,China;School of Computer Science,China West Normal University,Nanchong 637009,Sichuan,China)

机构地区：[1]南充职业技术学院电子信息工程系,四川南充637131 [2]西华师范大学计算机学院,四川南充637009

出　　处：《计算机应用与软件》2021年第10期304-309,349,共7页Computer Applications and Software

基　　金：四川省教育厅重点项目(14ZA0123);南充市科知局课题(19YFZJ0029)。

摘　　要：针对大数据中高维数据处理搜索空间不佳和计算成本较高的问题,提出一种基于聚类和二元蚂蚁系统(Binary Ant System,BAS)相结合的混合滤波器特征选择算法。该算法由线性二元蚂蚁系统、聚类和阻尼突变三种策略组成,可以有效地克服搜索空间和高维数据处理的困难,在聚类的类别之间和类内部具有全局和局部搜索能力。实验结果表明,与其他特征选择方法相比,该方法明显降低了计算复杂度,在多个评估指标中取得了较好的性能。Aiming at the problem of poor search space and high computing cost of high-dimensional data processing in big data,a hybrid filter feature selection algorithm based on clustering and improved binary ant system(BAS)is proposed.This algorithm was composed of linear binary ant system,clustering and damping mutation.It can effectively overcome the difficulties of search space and high-dimensional data processing.It has global and local search ability between clusters and within clusters.Experimental results show that compared with other feature selection methods,this method significantly reduces the computational complexity and achieves better performance in multiple evaluation indexes.

关 键 词：高维数据 特征选择 聚类 二元蚂蚁系统 阻尼突变 大数据 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]