Evaluation object extraction method of microblog comments involved in the case based on multi-head attention mechanism

Abstract

The invention relates to a method for extracting evaluation objects of microblog comments involved in a case based on a multi-head attention mechanism. First, the text of the microblog involved and the microblog comments are collected; the evaluation objects of the microblog comments involved are marked; the character vector of the microblog comments involved and the words are obtained. vector, noun part-of-speech-related part-of-speech vector, context vector and case element vector; concatenate the obtained character vector and word vector, and use the multi-head attention mechanism to obtain the correlation between the part-of-speech vector and the context vector, and calculate the largest case element and Similarity probability weight of part-of-speech vector; Dot multiplication of the obtained splicing vector, correlation weight and maximum similarity probability weight, and the obtained result is used as the input of the BiLSTM model, and then the final sequence label is obtained through the conditional random field CRF, that is, extraction result. The present invention achieves a good extraction effect, which is 1% and 1.9% higher than the baseline model in two typical microblog comment datasets involved in the case, respectively.

Description

technical field [0001] The invention relates to an evaluation object extraction method for microblog comments involved in a case based on a multi-head attention mechanism, and belongs to the technical field of data mining. Background technique [0002] With the advent of the self-media era, the vigorous development of new media in my country and the deepening of online public opinion, some of the key cases heard by the courts are rapidly fermenting on the Internet, causing widespread social controversy, and even affecting judicial trials, seriously affecting social stability. In this context, the country is also promoting the construction of smart courts. The judicial department can effectively grasp and correctly guide online public opinion in real time to help ensure that courts exercise judicial power independently and maintain social fairness and justice. As one of the social media for free comments by the general public, Weibo has a large number of users and rapid devel...

AI Technical Summary

Problems solved by technology

[0003] The present invention provides a method for extracting evaluation objects of microblog comments involved in the case based on a multi-head attention mechanism, which is used to solve the problem that the evaluation objects of public comments in the microblog comments involved in the case are not related to the case involved and the public is concerned about the microblog comments involved in the case. Inconsistencies in the expression of the evaluation object when commenting

Images