Extraction Method of Judicial Language Entities Based On Regular Expression

摘要

With the coming of the era of rule of law and intelligence, natural language processing technology plays a pivotal role. At present, a large number of unstructured judicial texts rely on manual processing and archiving. In order to make better use of them and achieve professional application, this paper proposes the goal of analyzing the structure of judgments, extracting the judicial language entities, and describing cases in the form of entity circulation map. As the text carrier of unstructured public events, the judicial document is of better standard format, finely crafted and easy processing, and becomes the research object of this paper. Through the survey of the development of named entity recognition technology, testing and contrasting the use of extraction tool, GATE, as well as considering the cost and effectiveness in the judicial field, this paper put forward a rule-based regular expression method for entity recognition. The scrapy crawler framework is used to obtain judgments classified from China Judgments Online website, so as to realize the task of analyzing the structure of judgments and extracting the judicial language entities.