一种高效的网页聚类方法

当前主流的搜索引擎主要是以与用户查询的相关度来顺序返回搜索结果的,用户往往需要花费较长的时间从结果列表中进行选择.为了解决这个问题,针对搜索引擎返回的标题和摘要信息,构造有向图表示,并在此基础上实现了一种高效的网页聚类原型系统(efficient web clustering system,EWCS).该系统将搜索引擎返回的结果按照一定的标准分类呈现给用户,用户选择感兴趣的类别进行浏览,从而较好地满足了用户对查询速度和准确度的需求.试验结果表明该算法具有一定的可行性和较高的准确率.

Effective approach to web clustering

Nowadays,almost all search engines return the search results in sequence according to the correlation rate with users' query,by which the users have to spend a lot of time on selecting the related ones from the list.To solve this problem,the titles and abstracts returned by search engines is used and then a new kind of directed graph representation is constructed,based on which the efficient web clustering system demo named EWCS is implemented.In the system,the results returned by search engines are presented to users after being classified based on certain standards and the users will choose to scan the ones they are interested in.By this way,the system satisfies the users comparatively better in the need of searching speed and accuracy.The experimental result show the system has certain feasibility and comparative accuracy.