Movie comment sentiment classification method and device based on graph neural network

Abstract

The invention discloses a movie comment sentiment classification method and device based on a graph neural network, and the method comprises the steps: constructing a graph data structure based on the specific grammar information of a text sentence and the window co-occurrence relation of text words, obtaining the adjacent matrixes of a text co-occurrence graph and a syntactic dependency graph respectively, and carrying out the classification of the movie comment sentiment. Further inputting the word embedded vector matrix and the adjacent matrix into a graph convolutional network module to obtain a weighted feature matrix, and performing pooling operation on the weighted feature matrix of the text co-occurrence graph and the weighted feature matrix of the syntactic dependency graph by adopting a mixed pooling module; and obtaining a feature vector of the text co-occurrence graph and a feature vector of the syntactic dependency graph, fusing the feature vectors, and inputting the fused feature vectors into a classifier to obtain a classification result. According to the method, the network representation obtained by learning not only contains the information of adjacent words, but also contains the information of relatively far words related to grammatical relationships, so that the accuracy of sentiment classification is improved.

Description

technical field [0001] The present application belongs to the technical field of natural language processing, and in particular relates to a method and device for sentiment classification of movie reviews based on a graph neural network. Background technique [0002] We are currently in an era of the Internet and big data. Massive data in various forms are generated every day, such as text, voice, video, etc., and text data accounts for a large part of it. With the rapid development of the cultural industry, the number of movies released each year is increasing, which brings various movie reviews. For example, there are a large number of movie reviews on Douban, Maoyan and other well-known movie websites. It becomes a key issue to dig out useful information from reviews, and an important step is to classify the sentiment of these movie reviews. [0003] Sentiment classification is a fundamental and important task in natural language processing. Early sentiment classificati...

AI Technical Summary

Problems solved by technology

However, such a method relies too much on manual feature extraction, which is relatively inefficient.

Images