首页> 外文学位 >Athens: A system for novel information discovery.
【24h】

Athens: A system for novel information discovery.

机译:雅典:一种新颖的信息发现系统。

获取原文
获取原文并翻译 | 示例

摘要

The serendipitous discovery of novel information in information repositories such as the Web is a challenging task. Existing retrieval tools, like search engines, can retrieve information about known topics. However, they cannot retrieve information about novel topics, that is topics whose existence is unknown to the user and which may be potentially interesting.; In this thesis, we propose Athens1, a system for discovering novel information in the Web. Athens retrieves novel information through a focused search. It comprises three fundamental components: closure to find the essential content of a set of search query terms; probe to create new contextualized queries for retrieving information of wider scope; and cluster to remove less relevant information. Given a set of initial query terms, the system repeats these steps twice to reach novel information relative to the initial query topic.; The Athens system is tested using several search queries and the results are qualitatively validated. The experimental results indicate that the system retrieves novel information regarding a broad array of topics that possess a high degree of relevance to the initial query topic. The results also demonstrate that the novel information retrieved shares a strong correlation with the query topic.; 1Bible, Acts 17:21: More than anything else, the people of Athens and the foreigners living there loved to hear and to talk about anything new.
机译:在诸如Web之类的信息存储库中偶然发现新信息是一项艰巨的任务。现有的检索工具(例如搜索引擎)可以检索有关已知主题的信息。但是,他们无法检索有关新颖主题的信息,这些主题的存在对于用户来说是未知的,并且可能是有趣的。在本文中,我们提出了Athens1,这是一种用于在网络中发现新颖信息的系统。雅典通过重点搜索来检索新颖的信息。它包含三个基本组件:闭包以查找一组搜索查询词的基本内容;探索创建新的上下文查询,以检索范围更广的信息;并聚集以删除不太相关的信息。给定一组初始查询词,系统将重复这些步骤两次,以获取有关初始查询主题的新颖信息。使用几个搜索查询对雅典系统进行了测试,并对结果进行了定性验证。实验结果表明,该系统检索有关与初始查询主题高度相关的广泛主题的新颖信息。结果还表明,检索到的新颖信息与查询主题具有很强的相关性。 1使徒行传17:21:比其他任何事情,雅典人民和住在那的外国人都喜欢听和谈论新事物。

著录项

  • 作者

    Vats, Nikhil.;

  • 作者单位

    Queen's University (Canada).;

  • 授予单位 Queen's University (Canada).;
  • 学科 Computer Science.
  • 学位 M.Sc.
  • 年度 2004
  • 页码 115 p.
  • 总页数 115
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

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