面向垂直搜索引擎的主题提取算法  

Topic Distillation Algorithm for Vertical Search Engine

作　　者：吕林涛[1] 陈丽萍[1] 周红芳[1] 

机构地区：[1]西安理工大学计算机科学与工程学院,西安710048

出　　处：《计算机工程》2009年第15期44-46,共3页Computer Engineering

基　　金：陕西省自然科学基础研究计划基金资助项目(2007F52)

摘　　要：针对HITS算法对所有链接分配相等权重导致产生主题漂移的问题,提出基于计算链接价值度及Web页面语义主题相似度对链接分配合理权重的HITS改进算法,突出链接重要度的差异。实验表明,该算法的主题相关度提高了13%～42%,且较好地避免了主题漂移问题,增强了采集信息的准确性,对垂直搜索引擎的研究有重要的理论和实际应用价值。HITS algorithm assigns same weight to links between Web pages, which results in topic drift. Aiming at this problem, a new improved HITS algorithm is proposed by assigning appropriate weights to links according to the link value and page semantic topic similarity. Improved HITS algorithm highlights the difference of links importance. Experiments show that the proposed algorithm can improve relevance ratio by 13%-42%. Furthermore, it can well avoid topic drift and improve the accuracy of information collection. It has important theoretical and practical values for vertical search engines research.

关 键 词：HITS算法 垂直搜索引擎 超链接 主题漂移 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]