Text sentiment classification model training method and device, computer equipment and medium

Abstract

The invention relates to the technical field of artificial intelligence, and provides a text sentiment classification model training method and device, computer equipment and a medium. The method comprises the following steps: obtaining a plurality of long texts, and segmenting each long text to obtain a plurality of text statements; calculating a TextRank value of each text statement in each longtext, and generating a text abstract for each long text according to the TextRank value; calculating an emotion score of each text statement in each text abstract; sorting a plurality of text statements in each text abstract according to the emotion scores, and generating a text data set according to the sorted text statements; and training a plurality of text data sets based on the pre-trainingmodel to obtain a text sentiment classification model. The sentiments of the long texts can be accurately classified, and the position information, the time sequence information and the semantic information of the original long texts are not lost.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a text emotion classification model training method, device, computer equipment and medium. Background technique [0002] Sentiment analysis of long medical texts is an important part of network public opinion monitoring, which can effectively distinguish negative information, enable managers to verify and explain negative information in a timely and effective manner, and monitor the outbreak of network public opinion at all times. [0003] At present, the word2vec method is mostly used to encode text to achieve text classification, or the pre-trained model BERT is used to achieve text classification. However, in the process of implementing the present invention, the inventor found that word2vec cannot solve problems such as polysemy and grammar. Although the pre-training model BERT can solve problems such as polysemy and grammar, it can only solve the problems of...

AI Technical Summary

Problems solved by technology

However, the inventor found in the process of implementing the present invention that word2vec cannot solve the problems of polysemy and grammar. Although the pre-training model BERT can solve the problems of polysemy and grammar, it can only solve the problems of text length less than 512 word text

It can be seen that the current text classification method has a good classification effect for short texts, but it is not suitable for the classification of long medical texts.

Images