Dependency Graph Enhanced Dual-transformer Structure for Aspect-based Sentiment Classification

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 6578–6588 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 6578 Dependency Graph Enhanced Dual-transformer Structure for Aspect-based Sentiment Classification Hao Tang1 , Donghong Ji1∗, Chenliang Li1 , Qiji Zhou1 1Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, China {tanghaopro,dhji,cllee,qiji.zhou}@whu.edu.cn Abstract Aspect-based sentiment classification is a pop- ular task aimed at identifying the correspond- ing emotion of a specific aspect. One sentence may contain various sentiments for different aspects. Many sophisticated methods such as attention mechanism and Convolutional Neural Networks (CNN) have been widely employed for handling this challenge. Re- cently, semantic dependency tree implemented by Graph Convolutional Networks (GCN) is introduced to describe the inner connection between aspects and the associated emotion words. But the improvement is limited due to the noise and instability of dependency trees. To this end, we propose a dependency graph enhanced dual-transformer network (named DGEDT) by jointly considering the flat repre- sentations learnt from Transformer and graph- based representations learnt from the corre- sponding dependency graph in an iterative interaction manner. Specifically, a dual- transformer structure is devised in DGEDT to support mutual reinforcement between the flat representation learning and graph-based repre- sentation learning. The idea is to allow the dependency graph to guide the representation learning of the transformer encoder and vice versa. The results on five datasets demonstrate that the proposed DGEDT outperforms all state-of-the-art alternatives with a large mar- gin. 1 Introduction Aspect-based or aspect-level sentiment classifica- tion is a popular task with the purpose of identify- ing the sentiment polarity of the given aspect (Yang et al., 2017; Zhang and Liu, 2017; Zeng et al., 2019). The goal is to predict the sentiment po- larity of a given pair (sentence, aspect). Aspects in our study are mostly noun phrases appearing in the ∗Corresponding author. input sentence. As shown in Figure 1, where the comment is about the laptop review, the sentiment polarities of two aspects battery life and memory are positive and negative, respectively. Giving a specific aspect is crucial for sentiment classification owing to the situation that one sentence sometimes contains several aspects, and these aspects may have different sentiment polarities. Modern neural methods such as Recurrent Neu- ral Networks (RNN), Convolutional Neural Net- works (CNN) (Dong et al., 2014; Vo and Zhang, 2015) have already been widely applied to aspect- based sentiment classification. Inspired by the work (Tang et al., 2016a) which demonstrates the importance of modeling the semantic connection between contextual words and aspects, RNN aug- mented by attention mechanism (Bahdanau et al., 2015; Luong et al., 2015; Xu et al., 2015) is widely utilized in recent methods for exploring the poten- tially relevant words with respect to the given as- pect (Yang et al., 2017; Zhang and Liu, 2017; Zeng et al., 2019; Wang et al., 2016). CNN based atten- tion methods (Xue and Li, 2018; Li et al., 2018) …

Content truncated to 500 words