Aiming at the problems of insufficient extraction of semantic and syntactic information and neglect of the interaction between the two,an aspect-level sentiment analysis model using the interactive attention mechanism to integrate semantic and syntactic information was proposed.The inter-aspect dependency was fused with local semantics to obtain comprehensive global semantic information,and global and local information was interacted to obtain deeper semantic information.The improved graph convolutional network was used to enhance the model's ability to extract contextual syntactic information.Multi-head interaction attention was used to complete interactions between aspect words and context,as well as the enhanced semantics and syntax.To verify the effectiveness of the model,experiments were carried out on the Laptop14,Restaurat14 and Twitter benchmark data-sets.Experimental results show that the proposed model achieves better performance than the comparison method.
维普
万方数据