基于关键词相关度的搜索引擎设计——以天津电大远程教育资源库为例  

Design of Keyword Search Model on Query-Likelihood——Based on Learning Resource Repository of Distance Education in Tianjin Open University

作　　者：周莹[1] 

机构地区：[1]天津广播电视大学,天津300191

出　　处：《天津电大学报》2011年第3期43-46,共4页Journal of Tianjin Radio and Television University

摘　　要：基于语义Web的关键词相关度的搜索模型,是通过计算搜索关键词在各类教学资源网页主要内容文档块中出现的条件概率,来描述该网页与搜索关键词的相关度,同时利用教学资源网页与资源库课程的归属关系,修正计算的相关度结果。该智能搜索系统能够有效地解决用户搜索资源库时返回过多不相关页面的现状,同时根据媒体、课程与关键词的相关度进行排序,使用户能够从系统返回的近似媒体簇中迅速得到有用信息,从而提高检索准确率。Query-Likelihood of keyword search model of computer based Semantic Web is to compute the condition probability of the keywords in major subject blocks of all kinds of learning pages on the Web,describing the likelihood between the web and the keyword,and meanwhile,it can revise the computed likelihood result by refering to the relationship between the learning resources on the Web and the curriculum in the Learning Resource Repository.This intelligent search system can solve the problem of opening too many web pages when the user returns after searching in the Resource Repository.Based on the order of media,curriculum,and keyword likelihood,the user can quickly search for the useful information from the related media groups of returned system.In this way the accuracy of search can be improved.

关 键 词：语义Web 关键词 相关度 搜索模型 快速检索 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]