Yimmon at SemEval-2019 Task 9: Suggestion Mining with Hybrid Augmented Approaches

Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), pages 1267–1271 Minneapolis, Minnesota, USA, June 6–7, 2019. ©2019 Association for Computational Linguistics 1267 Yimmon at SemEval-2019 Task 9: Suggestion Mining with Hybrid Augmented Approaches Yimeng Zhuang Samsung Research China - Beijing (SRC-B) ym.zhuang@samsung.com Abstract Suggestion mining task aims to extract tips, advice, and recommendations from unstruc- tured text. The task includes many chal- lenges, such as class imbalance, figurative ex- pressions, context dependency, and long and complex sentences. This paper gives a de- tailed system description of our submission in SemEval 2019 Task 9 Subtask A. We trans- fer Self-Attention Network (SAN), a success- ful model in machine reading comprehen- sion field, into this task. Our model con- centrates on modeling long-term dependency which is indispensable to parse long and com- plex sentences. Besides, we also adopt tech- niques, such as contextualized embedding, back-translation, and auxiliary loss, to aug- ment the system. Our model achieves a per- formance of F1=76.3, and rank 4th among 34 participating systems. Further ablation study shows that the techniques used in our system are beneficial to the performance. 1 Introduction Suggestion mining is a trending research domain that focuses on the extraction of extract tips, advice, and recommendations from unstructured text. To better recognize suggestions, instead of only matching feature words, one must have the ability to understand long and complex sentences. SemEval-2019 Task 9 provides the suggestion mining task (Negi et al., 2019). The task can be recognized as a text classification task, given a sentence collected from user feedback, participat- ing systems are required to give a binary output by marking it as suggestion or non-suggestion. To address this problem, we focus on solving long-term dependency on long and complex sen- tences. Consequently, we transfer Self-Attention Network (SAN), a successful model in machine reading comprehension field, in which long-term dependency is crucial, into this task. Furthermore, we also utilize multiple techniques to improve the suggestion mining system. 2 System description In this paper, we consider suggestion mining as a text classification task. Figure 1 gives an overview of our model. First, the input text is converted into word embeddings with linguistic features. Then, we use several stacked semantic encoders to gen- erate the hidden representations for each token. On top of that, a softmax output layer estimates the probability of the text being a suggestion. 2.1 Input encoding The input encoding layer is in charge of encoding each token of the input text to singular vectors. To- kenization is completed during preprocessing. In our work, we adopt WordPiece embedding (Wu et al., 2016) and feed it into a pretrained lan- guage model (LM) to generate contextualized em- beddings. Compared to context independent word vectors, such as widely used GloVe (Pennington et al., 2014), SGNS (Mikolov et al., 2013), con- textualized vectors show significant advantages in disambiguation and sentence modeling. Besides, well-pretrained language model also transfers ex- ternal knowledge to this task, full use of transfer learning is the key to the advance …

Content truncated to 500 words