Position-Aware Tagging for Aspect Sentiment Triplet Extraction

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pages 2339–2349, November 16–20, 2020. c⃝2020 Association for Computational Linguistics 2339 Position-Aware Tagging for Aspect Sentiment Triplet Extraction Lu Xu* 1, 2, Hao Li* 1, 3, Wei Lu1, Lidong Bing2 1 StatNLP Research Group, Singapore University of Technology and Design 2 DAMO Academy, Alibaba Group 3ByteDance xu lu@mymail.sutd.edu.sg, hao.li@bytedance.com luwei@sutd.edu.sg, l.bing@alibaba-inc.com Abstract Aspect Sentiment Triplet Extraction (ASTE) is the task of extracting the triplets of target entities, their associated sentiment, and opin- ion spans explaining the reason for the senti- ment. Existing research efforts mostly solve this problem using pipeline approaches, which break the triplet extraction process into several stages. Our observation is that the three el- ements within a triplet are highly related to each other, and this motivates us to build a joint model to extract such triplets using a se- quence tagging approach. However, how to ef- fectively design a tagging approach to extract the triplets that can capture the rich interac- tions among the elements is a challenging re- search question. In this work, we propose the first end-to-end model with a novel position- aware tagging scheme that is capable of jointly extracting the triplets. Our experimental re- sults on several existing datasets show that jointly capturing elements in the triplet using our approach leads to improved performance over the existing approaches. We also con- ducted extensive experiments to investigate the model effectiveness and robustness1. 1 Introduction Designing effective algorithms that are capable of automatically performing sentiment analysis and opinion mining is a challenging and important task in the field of natural language processing (Pang and Lee, 2008; Liu, 2010; Ortigosa et al., 2014; Smailovi´c et al., 2013; Li and Wu, 2010). Recently, Aspect-based Sentiment Analysis (Pontiki et al., 2014) or Targeted Sentiment Analysis (Mitchell et al., 2013) which focuses on extracting target ∗ Equal contribution. Lu Xu is under the Joint PhD Pro- gram between Alibaba and Singapore University of Technol- ogy and Design. The work was done when Hao Li was a PhD student in Singapore University of Technology and Design. 1We release our code at https://github.com/ xuuuluuu/Position-Aware-Tagging-for-ASTE 0 + food was so so but excited to see many vegan options Figure 1: ASTE with targets in bold in solid squares, their associated sentiment on top, and opinion spans in dashed boxes. The arc indicates connection between a target and the corresponding opinion span. phrases as well as the sentiment associated with each target, has been receiving much attention. In this work, we focus on a relatively new task – As- pect Sentiment Triplet Extraction (ASTE) proposed by Peng et al. (2019). Such a task is required to extract not only the targets and the sentiment mentioned above, but also the corresponding opin- ion spans expressing the sentiment for each target. Such three elements: a target, its sentiment and the corresponding opinion span, form a triplet to be extracted. Figure 1 presents an example sentence containing two targets in solid boxes. Each …

Content truncated to 500 words