Abstract: In the process of clustering the visitors’ travel path patterns, this paper achieves a more reasonable and accurate model from combining the content as well as the path a user visits, at the same time improves the former research in measuring the similarity of travel path pattern and presents an effective way to perform it. The result shows that the improved method decreases the number of the clustering result, and is much better in finding the potential user classes, which achieves the expectation of this paper.

Key words: Vector space model, Web content mining, Web usage mining, Fuzzy clustering

摘要： 在用户访问模式的聚类过程中引入页面的相似性因子，从用户访问的主要内容和访问路径两个方面来度量访问模式的相似性，针对以往对这种集成研究忽略的问题进行深入的探讨，提出了有效的解决方法，合理地降低了聚类结果的类别数目，能更准确地发现一个网站的潜在用户类。

关键词: 向量空间模型, Web内容挖掘, Web使用挖掘, 模糊聚类

CLC Number: 

• TP311