XJIPC OpenIR  > 多语种信息技术研究室
基于用户兴趣模型的个性化搜索引擎研究
郑洁
Subtype硕士
Thesis Advisor李晓
2009-06-03
Degree Grantor中国科学院研究生院
Place of Conferral北京
Degree Discipline计算机应用技术
Keyword搜索引擎 个性化 向量空间
Abstract随着互联网的不断发展和日益普及,网上的信息量也在爆炸性增长。如何在众多的网页中,寻找到自己需要的信息,是当今搜索引擎的研究重点。目前的搜索引擎存在诸如缺乏信息收集和信息检索的同步性、信息检索方式具有单一性、信息检索内容具有单一性和信息服务方式具有被动性等问题,也就是说大部分现有的搜索引擎不能提供针对用户的个性化服务,对于不同的用户,输入同一个关键词返回的结果是一样的。随着网页数量指数级增长,搜索引擎应该从如何找到更多的信息转换为如何找到更为准确、有用的信息,应该对用户提供针对性的服务。这就引出了个性化搜索引擎的概念,而有关用户建模技术已经成为个性化服务研究的关键技术。目前,个性化信息服务中主要有三种用户建模技术:用户手工定制、用户浏览网页时进行标注和评价和系统自动建模。前两种是系统被动学习,需要用户的参与,这样会增加用户的负担。后一种是系统自动学习,不会干扰用户的正常浏览,是比较优越的方法。但总的来说,系统自动建模技术还处于起步阶段,尚未形成完整的技术体系。 本文在研究各种个性化检索技术的基础上,采用用户手工定制和系统自动建模相结合的形式,利用向量空间模型的表示方法,采用基于个性化信息采集的个性化检索思想,研究并改进了一种基于用户兴趣模型的智能调整算法来实现个性化推荐服务。
Other AbstractAlong with the developing and prevalence of internet, the information of internet grow rapidly. How to find the needful information among the numerous web pages is the search engineer’s researching emphases. Now, most search engineers have the faults such as the synchronization of gathering pages and searching information, the oneness of modes of searching information, the oneness of content of searching information, the service of information being passive and so on. That is to say most search engineer don’t provide the special service to users. If different user input the same query words, the system will give the same results. Along the rapid growing of amounts of web pages, search engineer should transit from how to find more information to how to find more exact and useful information and provide the special service to different users. Then the concept of individuation search is been putted forward, and the technology of user modeling has been to the key way to realize the individuation search . Now, there are 3 modes of individuation search: user customizing by handwork, user labeling and apprising when browsing web pages and system modeling automatically. The former two modes need system study passively, need users take part in, so it will gain weight of users. The last mode need system study initiative, it will not disturb the user’s browsing. But, the system initiatively modeling is just now in research, haven’t form the whole technology system. This article combing the modes of user customizing and system modeling and using the vector space model and using the idea based individuation information collection to study the arithmetic based user interesting model to achieve the individuation commending service.
Document Type学位论文
Identifierhttp://ir.xjipc.cas.cn/handle/365002/3584
Collection多语种信息技术研究室
Affiliation中国科学院新疆理化技术研究所
Recommended Citation
GB/T 7714
郑洁. 基于用户兴趣模型的个性化搜索引擎研究[D]. 北京. 中国科学院研究生院,2009.
Files in This Item:
File Name/Size DocType Version Access License
郑洁硕士论文.pdf(870KB)学位论文 开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[郑洁]'s Articles
Baidu academic
Similar articles in Baidu academic
[郑洁]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[郑洁]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 郑洁硕士论文.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.