结合Web用户访问特点,针对Web用户访问路径聚类分析中普遍存在的对象类别不确定性现象进行了研究.结合模糊聚类和可能性聚类的特点,提出来一种新的用户访问路径的可能性模糊聚类算法.新方法通过定义相关的截集,自动地将对象分配到若干簇中,避免了人工干预,实现了交叉聚类的目的.新方法建立在leader聚类算法的框架上,只需要扫描数据集一遍使得算法效率大大提高.在标准数据集上的对比试验表明新算法不仅是有效的,而且效率较高.
A novel uncertain clustering method is proposed in this paper after taking into account the characteristics of users' browsing actions. Based on combination of the fuzzy clustering and the possibilistic clustering, a possibilistic fuzzy clustering algorithm based on web user access paths is proposed. A λ-cut set is defined to process the overlapping clusters adaptively. Considering the advantages of the leader algorithm in time efficiency, the framework of the leader algorithm is used here. The comparison of experimental results shows that the proposed algorithm is valid and efficient.