Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (1): 112-115.DOI: 10.3778/j.issn.1002-8331.2010.01.035

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Fuzzy K-prototypes clustering based on quantum genetic algorithm

YE Qi-ming1,LIANG Gen2   

  1. 1.School of Science,Maoming University,Maoming,Guangdong 525000,China
    2.Education Information and Technology Center,Maoming University,Maoming,Guangdong 525000,China
  • Received:2009-07-27 Revised:2009-10-09 Online:2010-01-01 Published:2010-01-01
  • Contact: YE Qi-ming

量子遗传算法的模糊K-prototypes聚类

叶奇明1,梁 根2   

  1. 1.茂名学院 理学院,广东 茂名 525000
    2.茂名学院 教育信息技术中心,广东 茂名 525000
  • 通讯作者: 叶奇明

Abstract: Cluster analysis is most widely used in data mining as a technology;it has important applications in many fields.Fuzzy h-prototypes algorithm is one of the most effective algorithms of cluster analysis,however,the problem of sensitive to initial value and vulnerable to the problem of local minimum exists.In order to overcome the shortcomings,a hybrid algorithm based on quantum genetic algorithm and FKP clustering algorithm is proposed.The quantum genetic algorithm is used to determine the initial cluster center FKP firstly,and then the results of quantum genetic algorithm clustering result is used as start value of follow-up FKP.Experimental results show that the algorithm has good convergence and stability,better than single use of FKP algorithms and related improved algorithms.

Key words: clustering algorithm, quantum genetic algorithm, fuzzy K-prototypes algorithm, numerical attributes, data mining

摘要: 聚类分析是数据挖掘中应用最多的一种技术,它在许多领域都有重要应用。模糊h-prototypes算法是当前聚类分析中最有效算法之一,但是存在对初始值敏感、容易陷入局部极小值的问题。为了克服该缺点,提出了一种基于量子遗传算法和FKP算法的混合聚类算法,首先利用量子遗传算法确定FKP的初始聚类中心,再将量子遗传算法聚类结果作为后续FKP算法的初始值。实验结果显示,算法具有良好的收敛性和稳定性,聚类效果优于单一使用FKP算法和相关改进的算法。

关键词: 聚类算法, 量子遗传算法, 模糊K-prototypes算法, 数值型属性, 数据挖掘

CLC Number: