基于StackOverflow的场景-解决方案知识图谱构建方法

SCENARIO SOLUTION KNOWLEDGE GRAPH EXTRACTION METHOD BASED ON STACKOVERFLOW

  • 摘要: 在软件问答社区中,帖子可能存在重复、滞后、信息冗余等问题。目前对StackOverflow的信息抽取工作以单个问题为研究对象,抽取准确度较差。为此,提出一种针对多问题场景抽取场景与解决方案的方法,设计并构建了场景-解决方案知识图谱。相比于已有工作,该方法能够对含有多个问题的场景进行信息抽取。实验证明,该方法抽取得到的结果更加准确。

     

    Abstract: In the software Q&A community, posts may have problems such as duplication, lag and information redundancy. At present, the information extraction methods of StackOverflow focus on single post, and the extraction accuracy is poor. Therefore, this paper proposes a method to extract scenarios and solutions for multiple problem scenarios, and designs and constructs a scenario-solution knowledge graph. Compared with the existing works, this method can extract information from multiple problem scenarios. The experimental results show that the results of this method are more accurate.

     

/

返回文章
返回