SCAN:Syntactic Knowledge and Commonsense Knowledge Adapter Based Network for Aspect-level Sentiment Classification

Aspect-level sentiment classification is a most pronounced approach, which is defined as an automated technique to extract significant information from a large number of texts. However, current research still has limitations in ALSC tasks (e.g. accuracy of dependency parsing and overlook of commonsense knowledge). In this work, we propose a syntactic knowledge and commonsense knowledge adapter based network, which deals with the position information, syntactic structure and external knowledge, respectively. The performance of our model is evaluated on the three benchmark datasets. Experimental results demonstrate that our model is a best alternative in ALSC tasks compared with the state-of-the-art methods.