A method and device for aspect-level sentiment analysis based on graph convolutional neural network

Abstract

An embodiment of the present invention provides an aspect-level sentiment analysis method and device based on a graph convolutional neural network. The method includes: acquiring a sentence to be subjected to aspect sentiment analysis, and aspect words in the sentence to be subjected to aspect sentiment analysis; Preprocess the sentences and aspect words to be subjected to aspect sentiment analysis, and obtain the input vector sequence and syntactic weighted graph corresponding to the sentence to be subjected to aspect sentiment analysis; input the input vector sequence and syntactic weighted graph into the pre-trained double graph convolutional neural network In the network, the sentiment analysis results corresponding to the aspect words are obtained. In the embodiment of the present invention, the double-graph convolutional neural network is used to not only focus on the syntactic features of the sentence, but also pay attention to the semantic features of the sentence, and extract the semantic-related features corresponding to the sentence, which makes up for the inaccurate defect of syntactic feature extraction for sentences that are not sensitive to syntax. Improve the accuracy of sentiment analysis results.

Description

technical field [0001] The present invention relates to the technical field of natural language processing, in particular to an aspect-level sentiment analysis method and device based on a graph convolutional neural network. Background technique [0002] ABSA (Aspect-based Sentiment Analysis, aspect-based sentiment analysis) is an entity-level fine-grained sentiment analysis task, which aims to judge the emotional polarity of a given aspect word in a sentence. Aspect-level sentiment analysis can more accurately identify the user's emotional attitude towards a specific aspect, rather than directly judging the emotional polarity at the sentence-level granularity. [0003] The existing aspect-level sentiment analysis uses the GCN (Graph Convolution Network, graph convolution network) based on the syntactic dependency tree. Specifically, the sentence to be subjected to aspect sentiment analysis is used as input information, and the pre-trained Glove word embedding is used to inp...

AI Technical Summary

Problems solved by technology

[0004] However, the inventors found that in the existing process of implementing aspect-level sentiment analysis using the GCN based on the syntactic dependency tree, the parsing results are inaccurate due to the syntactic analysis of the sentence by the syntactic analyzer, and when the sentence to be sentimentally analyzed When it is not sensitive to syntactic dependence, the undirected graph converted from the syntactic dependency tree parsed by the dependency syntactic analyzer is used as an input of GCN, which makes the sentiment analysis result obtained by GCN inaccurate

Images