关键词: 自然语言处理, 深度学习, 方面级情感分析, 交替注意力机制, 图卷积网络

Abstract: Aspect-level sentiment analysis aimed to predict the sentiment polarity of specific targets in given text. Aiming at the problem of ignoring the syntactic relationship between aspect words and context and reducing the attention difference caused by average pooling, an aspect-level sentiment analysis model based on Alternating Attention and Graph Convolutional Network (AA-GCN) was proposed. Firstly, the Bidirectional Long Short-Term Memory (Bi-LSTM) network was used to semantically model context and aspect words. Secondly, the GCN based on syntactic dependency tree was used to learn location information and dependencies, and the Alternating-Attention mechanism was used for multi-level interactive learning to adaptively adjust the attention to the target word. Finally, the final classification basis was obtained by splicing the corrected aspect features and context features. Compared with the Target-Dependent Graph Attention Network (TD-GAT), the accuracy rates of the proposed model on the four datasets increased by 2.67 %, 1.13 %, 1.27 %, 1.84 %, respectively, and the F1 values on the five datasets increased by 2.82 %, 0.98 %, 4.56 %, 4.89 %, 4.46 %, respectively, indicating the effectiveness of using syntactic relationships and increasing keyword attention.

Key words: natural language processing, deep learning, aspect-level sentiment analysis, alternating-attention mechanism, Graph Convolutional Network (GCN)