SIMGAT: A Sentiment Analysis Model Based on Graph Attention Mechanism

With the normalization of social media, the popularization of Chinese, how to effectively enable computers to recognize Chinese short-text messages is an important task for network public opinion management and control. Due to the complexity of social media, information between people will have mutual influence, that is, short-text information is interrelated and can be described as a form of graph data. This paper is based on the method of graph neural network for Chinese sentiment analysis, and proposes a method based on improved graph attention mechanism to learn the semantic and structural information between short-text content, and at the same time aggregate short-text information from the field, so as to effectively express the emotional context. The experimental results show that, compared with the existing methods, the graph-based sentiment analysis model is very effective, and the attention mechanism shows a better effect on the sentiment analysis task of short-text.