Enhancing Aspect Term Extraction with Soft Prototypes

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pages 2107–2117, November 16–20, 2020. c⃝2020 Association for Computational Linguistics 2107 Enhancing Aspect Term Extraction with Soft Prototypes Zhuang Chen, Tieyun Qian∗ School of Computer Science, Wuhan University, China {zhchen18, qty}@whu.edu.cn Abstract Aspect term extraction (ATE) aims to extract aspect terms from a review sentence that users have expressed opinions on. Existing studies mostly focus on designing neural sequence tag- gers to extract linguistic features from the to- ken level. However, since the aspect terms and context words usually exhibit long-tail distributions, these taggers often converge to an inferior state without enough sample ex- posure. In this paper, we propose to tackle this problem by correlating words with each other through soft prototypes. These proto- types, generated by a soft retrieval process, can introduce global knowledge from internal or external data and serve as the supporting ev- idence for discovering the aspect terms. Our proposed model is a general framework and can be combined with almost all sequence tag- gers. Experiments on four SemEval datasets show that our model boosts the performance of three typical ATE methods by a large mar- gin. 1 Introduction Aspect term extraction (ATE) is a fundamental sub- task in aspect-based sentiment analysis. Given a re- view sentence, ATE aims to extract all aspect terms that users have expressed opinions on. For exam- ple, from the review “The Bombay style bhelpuri is very palatable.”, ATE aims to extract “bhelpuri”. ATE has been widely studied in the last twenty years. Early researches are devoted to design rule- based (Popescu and Etzioni, 2005) and feature engineering-based (Li et al., 2010) methods. With the development of deep learning techniques, re- cent researches mostly regard ATE as a sequence labeling task and focus on developing various types of neural models (Liu et al., 2015; Xu et al., 2018; Ma et al., 2019) to generate a tag sequence for the review. *Corresponding author. � ��� ��� ��� ��� ��� ��� ��� � � � � � �� ����������� �������������� ����� ����� ����� ����� � � � �� �� �� �� ����������� (a) Aspect terms. � ��� ��� ��� ��� ��� ��� ��� � � � � � �� ����������� �������������� ����� ����� ����� ����� � ��� ��� ��� ���� ���� ���� ���� � � � � � �� ����������� �������������� ����� ����� ����� ����� (b) Context words. Figure 1: The distributions of aspect terms and context words in the training sets of four SemEval datasets. Though achieving impressive progress, current sequence taggers mentioned still face a serious chal- lenge: the taggers may converge to an inferior state due to the lack of samples for tail words. As shown in Figure 1, about 80% aspect terms and context words (i.e., non-aspect terms) appear no more than five times in the commonly-used SemEval datasets. Without enough sample exposure, neural models can hardly achieve an optimal performance (He et al., 2018; Chen and Qian, 2019). To tackle this challenge, correlating samples with …

Content truncated to 500 words