摘要
随着互联网上的信息日益增长,个性化的搜索需求越来越迫切,由于用户兴趣的不同和行为的差异,如何为不同的用户提供不同的检索结果成为一个具有挑战性的问题。首先对现有搜索引擎的个性化信息检索和查询扩展技术进行了分类总结,分析了它们各自的优缺点。然后提出了基于社会化标签的个性化查询词扩展方法。这些方法通过从用户所收藏的社会化标签或标签所对应的网页中提取出和用户查询词相关的词,来对用户的初始查询进行扩展。最后利用Delicious网站上的用户数据,对比研究了这几种个性化查询扩展算法。通过与Google进行对比分析实验,结果表明所提出的社会化标签的个性化查询词扩展方法能够较好地满足用户的个性化需求,检索结果比Google的检索结果更接近用户需求。
Personalized search is highly demanded along with the explosion of information on Internet. Because of the diversity of users＇ interests and behaviors, it is a great challenge to provide appropriate search results for different users. This paper reviews the existing query expansion techniques, and analyzes their advantages and disadvantages. Then four new personalized query expansion approaches based on social bookmarks are proposed： Tags analysis extension algorithm, tag cluster analysis extension algorithm, TF analysis query expansion algorithm and social tags query co-occurrence analysis extension algorithm. A variety of assessment methods and Google search results are compared. Experimental results show that the proposed social tag query co-occurrence analysis extension algorithm is better to meet the personalized needs of users.
出处
《计算机科学与探索》
CSCD
2010年第9期812-829,共18页
Journal of Frontiers of Computer Science and Technology
基金
国家自然科学基金No.60803037
60803036
国家高技术研究发展计划(863)No.2009AA01Z143
中央高校基本科研业务费专项资金No.HEUCFZ1010
HEUCF100602~~
