| 基于主动进化遗传算法的k-means聚类方法 |
| |
| 引用本文： | 刘园园.基于主动进化遗传算法的k-means聚类方法[J].青岛职业技术学院学报,2010,23(1):53-56. |
| |
| 作者姓名： | 刘园园 |
| |
| 作者单位： | 青岛职业技术学院,软件与服务外包学院,山东,青岛266555 |
| |
| 摘 要： | 由主动进化思想提出一种基于中心定位算子的遗传算法（GCOGA）。GCOGA算法通过对聚类中心的个数和选取进行指导,解决了常规k-means聚类方法对初始聚类中心的敏感性以及聚类结果与样本输入次序有关等问题。实验结果显示,该算法避免了k-means方法中对初始值敏感和容易陷入局部最优解的缺陷,使聚类更合理,效果更好。
|
| 关 键 词： | 遗传算法 主动进化 中心定位算子 k-means |
K-means Clustering Based on Genetic Algorithm of Active Evolution |
| |
| LIU Yuan-yuan.K-means Clustering Based on Genetic Algorithm of Active Evolution[J].Journal of Qingdao Vocational and Technical College,2010,23(1):53-56. |
| |
| Authors: | LIU Yuan-yuan |
| |
| Institution: | LIU+Yuan-yuan(Software+%26+Service+Outsourcing+School,Qingdao+Technical+College,Qingdao,Sh,ong+266555,China) |
| |
| Abstract: | A gene-center-oriented genetic algorithm(GCOGA) based on active evolution is proposed.By instructing the amount and selecting of cluster centers,GCOGA solves the problems such as sensitivity to the original clustering center and relevancy to the importing order of samples that exist in common k-means clustering algorithm.The experimental results show that GCOGA can successfully avoid the sensitivity to the original clustering center and the disadvantage of partial optimum solutions,so GCOGA is more rational... |
| |
| Keywords: | k-means |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
