期刊文献+

基于近邻信息和PSO算法的集成特征选取 被引量:9

Ensemble Feature Selection Method Based on Neighborhood Information and PSO Algorithm
下载PDF
导出
摘要 提出了一种新的PSO特征选取方法.以粒子对应特征组合的同类近邻样本和异类近邻样本间的距离关系作为类别可分性和粒子适应度函数.以适应度函数加权的群体历史最佳、粒子历史最佳和粒子邻域内最佳个体信息共同指导粒子运动方向,搜索类内紧密、类间分离的最佳特征组合;同时,利用加权集成方法对PSO特征选取方法进行集成,以提高特征选取方法的稳定性和鲁棒性.在5个高维数据集上的特征选取实验结果表明集成PSO特征选取方法的有效性和可行性. A new PSO algorithm is proposed in this paper for feature selection. Distances within the same class and between different classes are used as the index for distinguishing different classes, and thus can be used to construct the fit- ness function of particles in PSO. The direction of particles for searching optimal features which can result in close intra-class distance and far inter-class distance is determined by the current best solution of the particle and the optimal individual in particle neighborhood, weighted by the fitness function. Meanwhile, the PSO algorithm is aggregated by the weighted voting method to improve its stability and robustness. The experiment results on 5 high dimensional datasets show that the ensemble PSO algorithm is effective and feasible.
出处 《电子学报》 EI CAS CSCD 北大核心 2016年第4期995-1002,共8页 Acta Electronica Sinica
基金 国家自然科学基金(No.61475071 No.61173068 No.10172043) 教育部博士点基金(No.20093218110024) 江苏省自然科学基金青年基金(No.BK20141032) 国家质检总局科技项目(No.2013QK194) 安徽省自然科学基金(No.1608085QF157)
关键词 特征选取 PSO 集成方法 分类 feature selection PSO ( particle swarm optimization) ensemble method classification
  • 相关文献

参考文献22

  • 1Liu H, Sun J, Liu L, et al. Feature selection with dynamic mutual information[ J ]. Pattern Recognition, 2009,42 ( 7 ) : 1330 - 1339.
  • 2邹涛,张翠,田新广,张尔扬.概念级误用检测系统的认知能力研究[J].电子学报,2004,32(10):1694-1697. 被引量:1
  • 3李颖新,刘全金,阮晓钢.一种肿瘤基因表达数据的知识提取方法[J].电子学报,2004,32(9):1479-1482. 被引量:13
  • 4边肇祺.模式识别[M].北京:清华大学出版社,1987..
  • 5Zhang Daoqiang, Chen Songcan, Zhou Zhi-Hua. Constraint score.A new filter method for feature selection with pair- wise constraints[ J ]. Pattern Recognition, 2008,41 ( 5 ) : 1440 - 1451.
  • 6Guyon I, Weston J, Barnhil S, et al. Gene selection for cancer classification using support vector machines [ J]. Machine learning, 2002,46 ( 1 - 3 ) : 389 - 422.
  • 7王树林,王戟,陈火旺,李树涛,张波云.肿瘤信息基因启发式宽度优先搜索算法研究[J].计算机学报,2008,31(4):636-649. 被引量:17
  • 8Kennedy J, Eberhart R C. Particle swarm optimization[ A]. Proceedings of International Conference on Neutral Net- works IV[ C ]. Piscataway NJ : IEEE Service Center, 1995. 1942 - 1948.
  • 9朱大林,詹腾,张屹,郑小东.多策略差分进化的元胞多目标粒子群算法[J].电子学报,2014,42(9):1831-1838. 被引量:10
  • 10Kennedy J,Eberhart RC. A discrete binary version of theparticle swarm algorithm[ A]. Proceedings of IEEE Inter- national Conference on Systems, Man, and Cybernetics [C]. Washington: 1EEE, 1997. 4104 - 4109.

二级参考文献98

共引文献80

同被引文献40

引证文献9

二级引证文献63

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部