【摘要】 搜索引擎往往返回给用户一个包含大量文档片段的列表 ,用户从中筛选出自己所需要的文档。文中提出一种预取代理的方法 :对搜索引擎返回的结果进行聚类分析 ,使得用户以主题的方式来查看结果 ,满足用户搜索请求的个性化服务 ;同时对聚类进行评价 ,推测出用户可能感兴趣的文档 ,并将它们预取过来 ,从而减少网络延迟。更多还原

【Abstract】 Users of Web searching engines are often sifting through the long order list of document snippets returned by the engines. A pre-fetching agent based on search result is proposed in this paper. First,the pre-fetching agent uses clustering algorithm to classify the original document snippets. It can help the user view search result as the specific topics. At the same time,the pre-fetching agent can evaluate each cluster and predict the Web documents that the user is interest in. By pre-fetching these documents it can reduce Web latency.更多还原