计算机工程 ›› 2010, Vol. 36 ›› Issue (8): 185-187.doi: 10.3969/j.issn.1000-3428.2010.08.065

• 人工智能及识别技术 • 上一篇    下一篇

基于特征分析的粒子群优化聚类算法

邓 貌,鲁华祥,金小贤   

  1. (中国科学院半导体研究所人工神经网络实验室,北京 100083)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2010-04-20 发布日期:2010-04-20

Particle Swarm Optimization Clustering Algorithm Based on Feature Analysis

DENG Mao, LU Hua-xiang, JIN Xiao-xian   

  1. (Lab of Artificial Neural Networks, Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083)
  • Received:1900-01-01 Revised:1900-01-01 Online:2010-04-20 Published:2010-04-20

摘要: 为提高粒子群优化聚类算法的性能,结合特征分析相关方法,提出一种新的串联聚类算法KPCA-PSO,阐述算法的基本原理和实施方案。在特征分析过程中,以一种简单有效的特征值选择方法避免手动选择特征值的繁琐过程。以人工数据和实际数据测试算法性能,实验结果表明该方法具有较好的聚类效果。

关键词: 特征分析, 核主成分分析, 粒子群优化算法, 聚类

Abstract: Combined with the related method of feature analysis, a new cascading clustering method named KPCA-PSO is proposed to enhance the effect of Particle Swarm Optimization(PSO) clustering algorithm. Basic principles and detailed realization of the method are illustrated. In the process of feature analysis, a simple and effective feature selection method is put forward to avoid the boring manual feature selection. The method is evaluated on artificial data and real data, and experimental results show that it gains good clustering effect.

Key words: feature analysis, Kernel Principal Component Analysis(KPCA), Particle Swarm Optimization(PSO) algorithm, clustering

中图分类号: