【24h】

A Filtering Method based on Multi-attribute Information

机译:基于多属性信息的过滤方法

获取原文
获取原文并翻译 | 示例

摘要

With the development of the computer network technology, a large amount of information can be acquired via the network. Information filtering technology especially is effective to extract only information to satisfy the user's query. Conventional information filtering technologies for documents have turned their attentions to only single attribute information, such as the frequency of words in the document. However, network documents, such as E-mail and HTML documents, tagged with various kinds of information have increased recently. In this paper, we propose the method to filter the required information based on multi-attribute values in the document. And, its application to the E-mail filtering system is shown, where multi-attribute values correspond to "Sender", "Theme", "Type" and "Time limit" of each E-mail document. Prom the experimental results, it is found that the accuracy of this method is higher than conventional methods, which use the statistics of word occurrences.
机译:随着计算机网络技术的发展,可以通过网络获取大量信息。信息过滤技术尤其有效地仅提取信息以满足用户的查询。用于文档的常规信息过滤技术已将注意力转移到仅单个属性信息上,例如文档中单词的出现频率。但是,以各种信息标记的网络文档,例如电子邮件和HTML文档,最近有所增加。在本文中,我们提出了一种基于文档中多属性值过滤所需信息的方法。并且,示出了其在电子邮件过滤系统中的应用,其中多属性值对应于每个电子邮件文档的“发件人”,“主题”,“类型”和“时间限制”。验证了实验结果,发现该方法的准确性高于传统方法,后者使用了单词出现的统计数据。

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