基于词共现关系的检索结果知识关联研究

Knowledge Connection of Retrieval Results Based on Co-word Analysis

基于文献关键词的共现分析很难反映文献隐含知识的关联。文本将主题模型、关联规则、共词分析等方法相结合,探讨检索结果的知识关联问题。首先,采用主题模型抽取检索结果文献摘要中的主题,形成主题词集;然后分析主题词集的关联关系,选择高关联的主题词对进行词共现分析;最后挖掘检索结果文献的知识关联。与单纯的关键词进行共词分析相比较,本文的方法能够较好的揭示文献所记录知识之间的关联,有效的实现知识的归纳和总结。

It is difficult to reflect implicit knowledge connections in the literature based on a co-word analysis of the keywords. In this article, we combine a topic model, association rules, and co-word analysis to discuss the knowledge connection of the retrieval results. First, we use the topic model to discover the topic words in abstracts found in the literature, and build a topic word set. We then analyze the association among the topic words set, choose a high asso- ciation topic words set for a co-word analysis, and finally mine the knowledge connection in the retrieval results. Compared with a simple co-word analysis of the keywords, the method described in this paper can be used to discover the knowledge connection in the retrieval results better, and summarize the knowledge more effectively. Therefore, it has some important implications regarding how to find the implicit connections among different studies in the litera- ture.