Construction method of evidence chain relation model for judgment document

Abstract

The invention provides a construction method of an evidence chain relation model for a Chinese judgment document. The process of the construction method is shown in the figure and mainly comprises thesteps that text preprocessing is performed on the Chinese judgment document; an XML-formatted file obtained after processing is read, and an evidence chain body set and a fact set are extracted; a keyword set of each evidence chain body and a keyword set of each fact are obtained according to a 4W1H key element extraction strategy; linking in the evidence chain relation model is constructed by calculating key element relevancy to complete model construction; the constructed model is made into an Excel table; and a structured model is displayed as a visual model. Through the method, mainly fora construction task of the evidence chain relation model in argument evaluation of the legal judgment document, according to the characteristics of judgment document arguments and the characteristicsof court document vocabularies, the judgment document preprocessing method is improved, the key element extraction technology is optimized, the correct rate of calculation of linking in an evidence chain is increased, and the method can be effectively used for evaluating the fact argument level of the judgment document.

Description

Technical field [0001] The invention relates to a judgment document argumentation analysis technology, in particular to a method for constructing a judgment document-oriented evidence chain relationship model, and belongs to the technical field of natural language processing. Background technique [0002] The research on the reasoning technology of the judgment document is mainly based on the extraction and analysis of the evidence chain. It studies the natural language processing technology that follows the French language and the document reasoning (language with the characteristics of legal terms), and extracts the key paragraphs and evidence of the judgment document reasoning. Reversely construct the logical relationship between evidence chain and clerical reasoning, and draw a visualized logical diagram of evidence chain and clerical reasoning through computer programs. For the processing of historical big data, it is necessary to introduce big data processing technologies s...

AI Technical Summary

Problems solved by technology

[0005] The writing of evidence and factual information in judgment documents has certain norms, but each evidence chain relationship model has its own special features and there are many uncertainties. For example, the number of chain heads of evidence chain links cannot be estimated, so the existing The modeling method and unsupervised machine learning are applied to the construction of the model

Images