Emotion analysis method and system, storage medium and equipment

Abstract

The invention relates to an emotion analysis method and system. The method comprises: using hidden layer outputs of the context embedding matrix and the target word embedding matrix as calculation parameters; and calculating attention mechanism coefficients of the context and the target word based on an attention mechanism, and performing weighted summation on the context vector and the target word vector according to the attention mechanism coefficients to obtain a new context vector and a new target word vector. Compared with the prior art, the influence coefficient between the target word and the context is calculated by using the attention mechanism, the importance of different words in the target word is distinguished, and the accurate analysis of the emotion of the target word is realized.

Description

technical field [0001] The present invention relates to the field of natural language processing, in particular to an emotion analysis method, system, storage medium and equipment. Background technique [0002] Natural language processing (NLP) involves the interaction between computers and human natural language, which provides morphological, syntactic and semantic tools to transform stored text from raw data to useful information. Natural language processing has always been considered one of the core issues in artificial intelligence research, because language not only plays an important role in human intelligence, but also has the potential to greatly benefit human daily life. Since 1950, NLP technology has come a long way. The most common application in our life - search engine, is a tool that uses NLP technology to facilitate users to retrieve information. [0003] Sentiment analysis is an important research direction of NLP, and its purpose is to identify the emotion...

AI Technical Summary

Problems solved by technology

It can be seen that a review sentence may involve the subjective opinions of multiple target words, and the research purpose of specific target sentiment analysis is to mine the target words in the sentence and analyze their emotional polarity, which is more granular than sentence or document level sentiment analysis. More detailed, belonging to fine-grained sentiment analysis; since specific target sentiment analysis first needs to identify the target words in the comment sentence, and then analyze its emotional polarity, it is more complex than sentence or document level sentiment analysis tasks

[0004] When performing sentiment analysis on a specific target in the prior art, the sentence is often split to model, and the split words are modeled indiscriminately for the sentiment analysis of the specific target. Analysis tasks have different importance, simply splitting them and modeling them indifferently affects the efficiency and accuracy of the specific target sentiment analysis task

Images