基于量子遗传谱聚算法的聚类.doc

文档介绍：基于量子遗传谱聚算法的聚类
蒋勇I,谭怀亮2
(，湖南，株洲，412004;
，湖南，长沙，410082 )
(hunanlaojiang®163. com)
摘 要：主要核方法研究XML聚类，提出了一种改进的XML X档核聚类方法。该方法先对XML文档约简，以 频繁标签序列建立向量空间核的核矩阵，用高斯核函数求解初始聚类和聚类中心，然后用初始聚类中心构 造量子遗传算法的初始种群，通过量子遗传算法与核聚算法相结合求得全局最优解的聚类。为了验证本文 提出的算法，实验结果显示，使用该算法的聚类比改进的核聚算法、K-means等单一方法具有良好的收敛 性、稳定性和更高的全局最优。
关键词：XML文档；高斯核函数；核聚类算法；量子遗传算法；XML聚类 中图法分类号：TP301. 6 文献标识码：A
Clustering Based on Quantum Genetic Spectral Clustering Algorithm
JIANG Yong1,TAN Hui-Wang2
(\ .Department of Information and College of Hunan Chemical ,Zhuzhou Hunan 412004, Ch 加 a
2. School of Computer and Communication,Hunan University, Changsha Hunan ,410082, China)
Abstract: This paper mainly t arge ts on XML Clus tering with kernel met hods for patt ern analysis and the quantum gene tic algor it hm, A new method based on the quantum gene tic algor ithm and clustering algorithm was derived. To the XML documents el iminated, the vec tor space kernel' s kernel mat rix were genera ted with frequent-tag sequence , first solves the initial clustering and clustering center with the Gaussian kernel functions , then the quantum genetic algorithm's initial populations were constructed by the initial clustering center structure, clustering of the globally optimal solutions were obtained through it and ka-nel clustering algorithm . In order to confirm the algorithm which this article proposed, the exp