Criminal name prediction method and device

Abstract

The embodiment of the invention provides a criminal name prediction method and device. The method comprises the following steps: carrying out the preprocessing of a fact description text of a to-be-predicted case, and obtaining a plurality of word vectors; inputting the plurality of word vectors into a criminal name prediction network model in sequence to obtain a criminal name prediction result of the to-be-predicted case; the criminal name prediction network model is obtained by training based on a case description text in a criminal legal document sample set, an actual criminal name corresponding to the case and attribute information corresponding to the actual criminal name. According to the embodiment of the invention, the attribute information of the criminal names is added into themodel training process, so that the trained criminal name prediction model can effectively predict the criminal names with fewer samples, the criminal names which are easy to confuse can be better distinguished by utilizing the attribute information of the criminal names, and the accuracy of automatic criminal name prediction is improved.

Description

technical field [0001] Embodiments of the present invention relate to the fields of machine learning and natural language processing, and more specifically, to methods and devices for predicting criminal charges. Background technique [0002] Automatic charge prediction is to automatically determine the charge of the defendant in a criminal case based on the description of the facts of the case. For decades, many experts and scholars in the legal and computer fields have proposed different methods for the crime prediction task. In existing methods, relevant methods in machine learning are often used to predict crimes for shallow text features (words, phrases) or attribute features (dates, locations, etc.) described in facts in a case. [0003] In recent years, with the development of machine learning, especially deep learning, some new methods have been introduced into this field. We can regard crime prediction as a text classification problem based on factual descriptions...

AI Technical Summary

Problems solved by technology

However, after excluding some crimes that rarely occur in real life, there are still about two hundred crimes that appear less frequently

For this part of the crime, due to the small number of original cases, automatic crime prediction is often difficult to achieve good results

Therefore, there is still a lot of room for improvement in the effect of automatic crime prediction for crimes with a low frequency of occurrence and a small number of case samples.

At the same time, there are some similarities between automatic crime prediction and artificial determination of crimes, which will cause some confusion for some relatively similar crimes.

Images