基于模糊聚类的超链分析算法  

An Auto-Categorizing Algorithm for Web Pages Ranking

作　　者：严海兵[1] 崔志明[2] 

机构地区：[1]苏州科技学院信息资源与技术开发研究所,苏州215011 [2]苏州大学计算机科学与技术学院,苏州215006

出　　处：《情报学报》2007年第3期361-365,共5页Journal of the China Society for Scientific and Technical Information

基　　金：江苏省自然科学基金资助项目（BG2005019）.

摘　　要：基于关键字匹配的搜索引擎排序网页时仅仅考虑评价网页的重要性，而忽视分类；基于分类目录的搜索引擎很难动态分析Web信息。本文在分析它们不足的前提下，提出利用模糊聚类的方法对搜索引擎的检索结果进行动态分类，依据超链分析算法PageRank和Web文档隶属度相结合进行分类排序，并给出具有调节值的结合公式。实验证明，该算法能够更有效地满足用户的需要，提高检索效率。Search engines on string match take into account Web pages ranking by estimating the pages＇ authority, while ignoring Web pages categorizing. It is difficuh for search engines on directory categories to auto-categorize Web pages in the Internet. For these deficiency of displaying the results in the current search engines, this paper attempts to improve pages ranking algorithm on hyperlink analysis in another point of view. By integrating the traditional hyperlink analysis and fuzzy clustering theory, the authors present an algorithm based on Auto-Categorizing, thereby a coefficient is got to adjust to the practical conditions. And the authors＇ experiment shows that this algorithm has better performance.

关 键 词：模糊聚类 超链分析 搜索引擎 WEB聚类 

分 类 号：TP393.092[自动化与计算机技术—计算机应用技术]