Information processing method and related equipment

Abstract

The embodiment of the invention provides an information processing method and related equipment, which can analyze element labels which better accord with judgment documents. The method comprises thesteps of obtaining a target judicial document, wherein the target judicial document is a judicial document of a to-be-analyzed element label; determining a target annotation sentence corresponding tothe target judicial document and a target element label corresponding to the target annotation sentence; performing word segmentation and feature vectorization processing on the target annotated sentence and the target element label to obtain a target sentence vector corresponding to the target annotated sentence and a target element vector corresponding to the target element label; and inputtingthe target sentence vector and the target element vector into a preset retrieval model to obtain an element label corresponding to the target judicial document.

Description

technical field [0001] The invention relates to the field of information processing, in particular to an information processing method and related equipment. Background technique [0002] Artificial intelligence + justice is in full swing. How to use the 40 million documents published by the Judgment Documents Network to provide effective support for the public security and law enforcement, reduce the repetitive work of judicial personnel, improve work efficiency, and promote judicial transparency and fairness. very meaningful thing. In how to use judicial documents, the structuring of documents is a very important process, and element label analysis of documents is the basis of judicial structuring. This work can make documents appear in the form of element labels, making legal workers Instead of reading the entire document from cover to cover to summarize the main issues covered by the document, simply browse the document's Elements tab. After analyzing the element tags ...

AI Technical Summary

Problems solved by technology

[0004] The disadvantage of the former is that for judgment documents, especially civil documents, the summary rules are completely dependent on manpower, and each regular expression is very complicated, especially for civil documents, there will be a variety of completely different descriptions for the same element label, and the summary rules It is very complicated, and the effect is often not good, and it is difficult to balance the accuracy and recall

The disadvantage of the latter is that to train a deep learning model for multi-class classification, the required labeling data needs to be very considerable, and in general, there will be serious data skew problems, that is, a small number of element labels can be labeled and accumulated in a small number of documents A large amount of data, and a large amount of data labeling a lot of data can only accumulate a small amount of data, the serious skew of the data will lead to the effect of the model is not very good

Images