A Multi-task Learning Framework for Opinion Triplet Extraction

Findings of the Association for Computational Linguistics: EMNLP 2020, pages 819–828 November 16 - 20, 2020. c⃝2020 Association for Computational Linguistics 819 A Multi-task Learning Framework for Opinion Triplet Extraction Chen Zhang1, Qiuchi Li2, Dawei Song1∗, Benyou Wang2 1 Beijing Institute of Technology, Beijing, China. 2 University of Padova, Padova, Italy. {czhang,dwsong}@bit.edu.cn, {qiuchili,wang}@dei.unipd.it Abstract The state-of-the-art Aspect-based Sentiment Analysis (ABSA) approaches are mainly based on either detecting aspect terms and their corresponding sentiment polarities, or co- extracting aspect and opinion terms. How- ever, the extraction of aspect-sentiment pairs lacks opinion terms as a reference, while co- extraction of aspect and opinion terms would not lead to meaningful pairs without deter- mining their sentiment dependencies. To ad- dress the issue, we present a novel view of ABSA as an opinion triplet extraction task, and propose a multi-task learning framework to jointly extract aspect terms and opinion terms, and simultaneously parses sentiment dependencies between them with a biaffine scorer. At inference phase, the extraction of triplets is facilitated by a triplet decoding method based on the above outputs. We evalu- ate the proposed framework on four SemEval benchmarks for ASBA. The results demon- strate that our approach significantly outper- forms a range of strong baselines and state-of- the-art approaches.1 1 Introduction Aspect-based sentiment analysis (ABSA), also termed as Target-based Sentiment Analysis in some literature (Liu, 2012), is a fine-grained sentiment analysis task. It is usually formulated as detecting aspect terms and sentiments expressed in a sen- tence towards the aspects (Li et al., 2019; He et al., 2019; Luo et al., 2019; Hu et al., 2019). This type of formulation is referred to as aspect-sentiment pair extraction. Meanwhile, there exists another type of approach to ABSA, referred to as aspect- opinion co-extraction, which focuses on jointly deriving aspect terms (a.k.a. opinion targets) and ∗Dawei Song is the corresponding author. 1Code and datasets for reproduction are available at https://github.com/GeneZC/OTE-MTL. Example sentence: The atmosphere is attractive , but a little uncomfortable . Aspect-sentiment pair extraction : [(atmosphere, positive), (atmosphere, negative)] Aspect-opinion co-extraction : [atmosphere, attractive, uncomfortable] Opinion triplet extraction : [(atmosphere, attractive, positive), (atmosphere, uncomfortable, negative)] Figure 1: Differences among aspect-sentiment pair extraction, aspect-opinion co-extraction, and opinion triplet extraction. Words in blue are aspect terms. Words in red are opinion terms. [ ] denotes a set of extracted patterns, and ( ) denotes an extracted pattern. opinion terms (a.k.a. opinion expressions) from sentences, yet without figuring out their sentiment dependencies (Wang et al., 2017; Li et al., 2018b). The compelling performances of both directions il- lustrate a strong dependency between aspect terms, opinion terms and the expressed sentiments. This motivates us to put forward a new perspec- tive for ABSA as joint extraction of aspect terms, opinion terms and sentiment polarities,2 in short opinion triplet extraction. An illustrative exam- ple of differences among aspect-sentiment pair ex- traction, aspect-opinion co-extraction, and opinion triplet extraction is given in Figure 1. Opinion triplet extraction can be viewed as an integration of aspect-sentiment pair extraction and aspect-opinion co-extraction, …

Content truncated to 500 words