模糊聚类的最大树算法在Web页面分类中的应用

Application of the Maximum Tree Algorithm for Fuzzy Clustering in the Clustering of Web Page

通过Web日志中记录的客户对Web页面的访问情况建立Web页面的用户访问矩阵,在此基础上构造模糊相似矩阵,根据模糊相似矩阵由最大树算法进行聚类。分析和算例表明,通过模糊相似矩阵进行聚类避免了构造模糊等价矩阵的大计算量,具有简单、快捷,适合处理高维数据的特点。

User access matrix is constructed with users accessing to the Web pages recorded in Web logs,the fuzzy similar matrix is obtained based on the user access matrix,finally,the clustering is performed according to the fuzzy similar matrix with the maximum tree algorithm.Analysis and example given show that the clustering with the fuzzy similar matrix can reduce the calculation volume comparing with constructing the fuzzy equivalence matrix,and it is simple,fast,suitable to deal with high-dimensional data.