...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of digital information management >Dynamic tree based classification of Web Queries using B-tree and Simple Ordinal classification algorithm
【24h】

Dynamic tree based classification of Web Queries using B-tree and Simple Ordinal classification algorithm

机译：使用B树和简单序数分类算法的基于动态树的Web查询分类

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Queries submitted by users to search engines might be ambiguous, concise and their meaning may change over time. Web query classification is emphasized by various search engines nowadays due to the increase in the size of the web, as millions of web pages are added to it every day. Some of the current Information Retrieval (IR) systems like Library Online Public Access Catalog (OPAC), dialog system and numerous web search engines need classical Boolean approaches in addition to the current supervised methods. Document retrieval is a process which mainly involves retrieval of relevant documents for user queries and matching the results using efficient algorithms like 'page rank' and 'learn to rank' algorithms. In this paper, to retrieve more relevant documents against user query by reducing non-relevant documents, we proposed a tree based classification of web queries using Simple Ordinal Classification (SOC) and navigation of search keywords is performed dynamically with search session time of users. This method reduces the retrieval of most of the non-relevant documents and navigation cost using efficient B-tree data structure. It provides documents that match all keywords present in the user query and best resulting web pages for users with different categories of interest. We built a prototype application to evaluate the proposed approach. Our experimental results revealed that SOC has significant performance improvement with existing approaches.
机译：用户向搜索引擎提交的查询可能是模棱两可的，简洁的，其含义可能会随时间而改变。由于网络规模的增加，当今各种搜索引擎都在强调网络查询的分类，因为每天都有数以百万计的网页被添加到其中。当前的某些信息检索（IR）系统，例如图书馆在线公共访问目录（OPAC），对话系统和许多Web搜索引擎，除了当前受监督的方法之外，还需要经典的布尔方法。文档检索是一个过程，主要涉及检索相关文档以供用户查询，并使用有效的算法（例如“页面排名”和“学习排名”算法）匹配结果。在本文中，为了通过减少不相关的文档来检索针对用户查询的更多相关文档，我们提出了一种使用简单序号分类（SOC）的基于树的Web查询分类，并且随着用户的搜索会话时间动态地执行搜索关键字的导航。这种方法使用有效的B树数据结构减少了大多数不相关文档的检索和导航成本。它为符合不同兴趣类别的用户提供了与用户查询中存在的所有关键字相匹配的文档以及最佳结果的网页。我们构建了一个原型应用程序来评估建议的方法。我们的实验结果表明，SOC与现有方法相比具有显着的性能改进。

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