计算机应用研究2018，Vol.35Issue(3)：680-684,5.DOI:10.3969/j.issn.1001-3695.2018.03.009
一种基于抗原软子空间聚类的否定选择算法
Improved negative selection algorithm based on antigen soft subspace clustering
摘要
Abstract
Negative selection algorithm (NSA) is an important method of detector-generation.Traditional NSAs ignored the difference of key characteristic and redundant characteristic of different kinds of antigens in the process of affinity-computing,which led to the poor performance.To solve this problem,this paper proposed an improved negative selection algorithm based on antigen soft subspace clustering(ASSC-NSA).First,by utilizing the antigen soft subspace clustering algorithm,ASSC-NSA found out all key characteristics and their weights of different kinds of antigens.Then,using the key characteristics to guide the detectors generation,thus it could eliminate the adverse influence of redundant characteristics and improve the detection rate.Compared with classical NSAs,the experimental result on BCW and KDDCup data set shows that ASSC-NSA improves the detection rate significantly with the similar false alarm rate.关键词
否定选择算法/软子空间聚类/异常检测Key words
negative selection algorithm/soft subspace clustering/anomaly detection分类
信息技术与安全科学引用本文复制引用
刘正军,高江锦,杨韬..一种基于抗原软子空间聚类的否定选择算法[J].计算机应用研究,2018,35(3):680-684,5.基金项目
国家自然科学基金资助项目(61572334) （61572334）
国家重点研发计划资助项目(2016YFB0800604) （2016YFB0800604）
南充市应用技术研究与开发资金资助项目(16YFZJ0011) （16YFZJ0011）
