Semantic emotion analysis method, device, and equipment and storage medium

Abstract

The invention relates to the field of intelligent decision in artificial intelligence, and discloses a semantic emotion analysis method, which comprises the steps of obtaining an importance metric value corresponding to each word in a to-be-analyzed statement; obtaining an implicit expression corresponding to the to-be-analyzed statement through two parallel running recurrent neural network modelsaccording to the importance metric value corresponding to each word in the to-be-analyzed statement; inputting the implicit expression corresponding to the to-be-analyzed statement and the sentence label corresponding to the to-be-analyzed statement into a semantic sentiment analysis classifier; and receiving an emotion analysis classification result of the semantic emotion analysis classifier for the to-be-analyzed statement. By introducing a self-attention mechanism, the importance of each word in the sentence is quantified through an importance measurement value, and then the meaning of the current word in the whole sentence is obtained according to the position of the important word, so that each word in the sentence and the corresponding importance measurement value thereof are fusedin the hidden state of the finally output whole sentence.

Description

technical field [0001] This application relates to the field of intelligent decision-making in artificial intelligence, in particular to a method, device, device and storage medium for analyzing semantic emotion. Background technique [0002] The earliest word-building model adopted was the bag-of-words model. The bag-of-words model regards a sentence as a simple collection of words, and combines them into a complete sentence through simple vector operations. With the development of deep learning, the application of neural networks is becoming more and more popular. RNN, as a sequential model, treats text as a sequence of words, which can effectively capture the relationship between time series variables, but the sequential model cannot distinguish sentence structures. The grammatical relationship in the sentence cannot determine the importance of each word in the sentence, which is not conducive to understanding the key points of the entire sentence, so it cannot recognize ...

AI Technical Summary

Problems solved by technology

[0003] The main purpose of this application is to provide a method for analyzing semantic sentiment, aiming at solving the technical problem of not being able to capture the nonlinear dependencies between common words in natural language and obtaining the sentiment features of sentences

Images