【24h】

Bipartite graph-based keyword query results recommendation

机译:基于二分图的关键词查询结果推荐

获取原文
获取原文并翻译 | 示例

摘要

Relational database is widely used in people's daily life and productions. Keyword search in relational database makes people search structured information from database as using search engine. So it has become a research hotspot. In recent years, there are a lot of research works about how to use historical query information to improve query efficiency. The existing works can be divided into two main aspects: query expansion and query recommendation. These two aspects have common feature, that is, they both use the historical words to improve query efficiency. If the results of historical query can be directly recommended to the current query, the query efficiency will be improved. But there are not any works using recommended model of recommended system to help keyword search. We have found that recommended model can be used to improve efficiency. We are the first one to use bipartite graph model, and propose query result recommendation algorithm, namely, Query_base algorithm and Distance_base algorithm. Finally this paper analyzes the factors which affect the efficiency of the algorithms, and verified the performance and higher accuracy of the recommendation algorithms by experiments.
机译:关系数据库广泛应用于人们的日常生活和生产中。关系数据库中的关键字搜索使人们像使用搜索引擎一样从数据库中搜索结构化信息。因此,它已经成为研究的热点。近年来,关于如何使用历史查询信息来提高查询效率的研究很多。现有的作品可以分为两个主要方面:查询扩展和查询推荐。这两个方面具有共同的特征,即它们都使用历史单词来提高查询效率。如果可以将历史查询的结果直接推荐给当前查询,则查询效率将得到提高。但是,没有使用推荐系统的推荐模型来帮助关键词搜索的作品。我们发现推荐模型可以用来提高效率。我们是第一个使用二部图模型的人,并提出了查询结果推荐算法,即Query_base算法和Distance_base算法。最后,本文分析了影响算法效率的因素,并通过实验验证了推荐算法的性能和较高的准确性。

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