...
【24h】

Rating News Documents for Similarity

机译:对新闻文件进行相似性评级

获取原文
获取原文并翻译 | 示例
   

获取外文期刊封面封底 >>

       

摘要

Electronic news has long held the promise of personal- ized and dynamic delivery of current event news items, particularly for web users. Although electronic versions of print news are now widely available, the personaliza- tion of that delivery has not yet been accomplished. In this paper, we present a methodology of associating news documents based on the extraction of feature phrases, where feature phrases identify dates, locations, people, and organizations. A news representation is cre- ated form these feature phrases to define news objects that can then be compared an ranked to find related news items. Unlike traditional information retrieval, we are much more interested in precision than recall. That is, the user would like to see one or more specifically related articles, rather than all somewhat related arti- cles. The algorithm is designed to work interactively with the user suing regular web browsers as the interface.
机译:电子新闻长期以来一直寄希望于个性化和动态地传递时事新闻,特别是对于网络用户。尽管现在印刷新闻的电子版本已广泛使用,但尚未完成该交付的个性化设置。在本文中,我们提出了一种基于特征短语提取的新闻文档关联方法,其中特征短语标识日期,位置,人员和组织。从这些特征短语创建新闻表示,以定义新闻对象,然后可以对这些新闻对象进行比较以找到相关的新闻项目。与传统的信息检索不同,我们对准确性的兴趣远胜于回忆。也就是说,用户希望看到一个或多个特定相关的文章,而不是所有有点相关的文章。该算法旨在与使用常规Web浏览器作为界面的用户交互工作。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号