DigNet: Digging Clues from Local-Global Interactive Graph for Aspect-level Sentiment Classification

In aspect-level sentiment classification (ASC), state-of-the-art models encode either syntax graph or relation graph to capture the local syntactic information or global relational information. Despite the advantages of syntax and relation graphs, they have respective shortages which are neglected, limiting the representation power in the graph modeling process. To resolve their limitations, we design a novel local-global interactive graph, which marries their advantages by stitching the two graphs via interactive edges. To model this local-global interactive graph, we propose a novel neural network termed DigNet, whose core module is the stacked local-global interactive (LGI) layers performing two processes: intra-graph message passing and cross-graph message passing. In this way, the local syntactic and global relational information can be reconciled as a whole in understanding the aspect-level sentiment. Concretely, we design two variants of local-global interactive graphs with different kinds of interactive edges and three variants of LGI layers. We conduct experiments on several public benchmark datasets and the results show that we outperform previous best scores by 3%, 2.32%, and 6.33% in terms of Macro-F1 on Lap14, Res14, and Res15 datasets, respectively, confirming the effectiveness and superiority of the proposed local-global interactive graph and DigNet.