Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence

Proceedings of NAACL-HLT 2019, pages 380–385 Minneapolis, Minnesota, June 2 - June 7, 2019. c⃝2019 Association for Computational Linguistics 380 Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence Chi Sun, Luyao Huang, Xipeng Qiu∗ Shanghai Key Laboratory of Intelligent Information Processing, Fudan University School of Computer Science, Fudan University 825 Zhangheng Road, Shanghai, China {sunc17,lyhuang18,xpqiu}@fudan.edu.cn Abstract Aspect-based sentiment analysis (ABSA), which aims to identify fine-grained opinion polarity towards a specific aspect, is a chal- lenging subtask of sentiment analysis (SA). In this paper, we construct an auxiliary sen- tence from the aspect and convert ABSA to a sentence-pair classification task, such as ques- tion answering (QA) and natural language in- ference (NLI). We fine-tune the pre-trained model from BERT and achieve new state-of- the-art results on SentiHood and SemEval- 2014 Task 4 datasets. The source codes are available at https://github.com/ HSLCY/ABSA-BERT-pair. 1 Introduction Sentiment analysis (SA) is an important task in natural language processing. It solves the com- putational processing of opinions, emotions, and subjectivity - sentiment is collected, analyzed and summarized. It has received much attention not only in academia but also in industry, provid- ing real-time feedback through online reviews on websites such as Amazon, which can take advan- tage of customers’ opinions on specific products or services. The underlying assumption of this task is that the entire text has an overall polarity. However, the users’ comments may contain dif- ferent aspects, such as: “This book is a hardcover version, but the price is a bit high.” The polarity in ‘appearance’ is positive, and the polarity regarding ‘price’ is negative. Aspect-based sentiment analy- sis (ABSA) (Jo and Oh, 2011; Pontiki et al., 2014, 2015, 2016) aims to identify fine-grained polarity towards a specific aspect. This task allows users to evaluate aggregated sentiments for each aspect of a given product or service and gain a more granu- lar understanding of their quality. ∗Corresponding author. Both SA and ABSA are sentence-level or document-level tasks, but one comment may re- fer to more than one object, and sentence-level tasks cannot handle sentences with multiple tar- gets. Therefore, Saeidi et al. (2016) introduce the task of targeted aspect-based sentiment analy- sis (TABSA), which aims to identify fine-grained opinion polarity towards a specific aspect associ- ated with a given target. The task can be divided into two steps: (1) the first step is to determine the aspects associated with each target; (2) the second step is to resolve the polarity of aspects to a given target. The earliest work on (T)ABSA relied heav- ily on feature engineering (Wagner et al., 2014; Kiritchenko et al., 2014), and subsequent neu- ral network-based methods (Nguyen and Shirai, 2015; Wang et al., 2016; Tang et al., 2015, 2016; Wang et al., 2017) achieved higher accuracy. Re- cently, Ma et al. (2018) incorporate useful com- monsense knowledge into a deep neural network to further enhance the result of the model. Liu et al. (2018) optimize the memory network and apply it to their model to better …

Content truncated to 500 words