摘要
为了克服传统的模糊K-Modes算法分类正确率低、收敛速度慢的缺点,文中将免疫遗传算法应用到聚类分析中,提出了一种基于模糊K-Modes和免疫遗传算法的聚类算法。通过引入免疫算子,不仅提高了收敛速度,而且避免了陷于局部极小,从而能较快地收敛到全局最优解。免疫算子包括抽取疫苗、接种疫苗和选择疫苗。实验结果证明,此算法具有较好的聚类效果,且稳定性强。
To overcome the shortcomings of the low clustering correctness and slow convergent speed in the basic fuzzy K - Modes clustering algorithm,the immune genetic algorithrn is introduced into the cluster analysis. Cluster analysis based on fuzzy K - Modes and immune genetic algorithm is proposed. Both the analytical and experimental studies indicate that this method is faster and more efficient to converge upon the optimal value. Immune operators including vaccine extraction, vaccination and vaccine selection. The experimental result shows that this improved algorithm is effective and steady contrast with basic fuzzy K - Modes.
出处
《计算机技术与发展》
2009年第2期151-153,共3页
Computer Technology and Development
基金
安徽省自然科学研究项目(2005kj001)
