面向类案检索的事理图谱构建方法研究

Research on the Construction Method of Affair Atlas Oriented to Class Case Retrieval

围绕司法案件裁判文书中词语的语义信息关系以及事件的信息表现特征,针对法院类案检索系统通过关键词匹配无法准确获取案件信息特征的问题,提出了一个基于自然语言处理技术的案件事理图谱构建方法。该方法采用基于信息抽取的裁判文书预处理技术、分布式词表示的文书语义特征提取技术、语义句法技术以及聚类的图谱触发词拓展技术联合构建案件事理图谱,深度挖掘隐藏在司法裁判文书之下的案件信息特征,验证并实现了一个高效、可靠、可扩展的事理图谱的构建方法。

Focusing on the semantic information relationship of the words in the judgment documents of judicial cases and the information performance characteristics of the event,in order to solve the problem that the court case retrieval system cannot accurately obtain the information characteristics of the case through keyword matching,a case affair map construction based on natural language processing technology is proposed method.This method uses information extraction-based judgment document preprocessing technology,distributed word representation document semantic feature extraction technology,semantic syntax technology,and clustering map trigger word expansion technology to jointly construct a case affair map,deep mining hidden in the judicial judgment document The characteristics of case information under the following verifies and realizes an efficient,reliable,and scalable judicial case rational map model.