Aspect Sentiment Classification Towards Question-Answering with Reinforced Bidirectional Attention Network

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3548–3557 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 3548 Aspect Sentiment Classification Towards Question-Answering with Reinforced Bidirectional Attention Network Jingjing Wang1, Changlong Sun2, Shoushan Li1,∗, Xiaozhong Liu2, Min Zhang1, Luo Si2, Guodong Zhou1 1School of Computer Science and Technology, Soochow University, China 2Alibaba Group, China djingwang@gmail.com, {lishoushan, minzhang, gdzhou}@suda.edu.cn, {changlong.scl, xiaozhong.lxz, luo.si}@alibaba-inc.com Abstract In the literature, existing studies on aspect sen- timent classification (ASC) focus on individ- ual non-interactive reviews. This paper ex- tends the research to interactive reviews and proposes a new research task, namely Aspect Sentiment Classification towards Question- Answering (ASC-QA), for real-world appli- cations. This new task aims to predict sen- timent polarities for specific aspects from in- teractive QA style reviews. In particular, a high-quality annotated corpus is constructed for ASC-QA to facilitate corresponding re- search. On this basis, a Reinforced Bidirec- tional Attention Network (RBAN) approach is proposed to address two inherent challenges in ASC-QA, i.e., semantic matching between question and answer, and data noise. Experi- mental results demonstrate the great advantage of the proposed approach to ASC-QA against several state-of-the-art baselines. 1 Introduction As a fine-grained sentiment analysis task, Aspect Sentiment Classification (ASC) aims to predict sentiment polarities (e.g., positive, negative, neu- tral) towards given particular aspects from a text and has been drawing more and more interests in natural language processing and computational linguistics over the past few years (Jiang et al., 2011; Tang et al., 2016b; Wang et al., 2018a). However, most of the existing studies on ASC fo- cus on individual non-interactive reviews, such as customer reviews (Pontiki et al., 2014) and tweets (Mitchell et al., 2013; Vo and Zhang, 2015; Dong et al., 2014). For example, in a customer re- view “The food is delicious, but ambience is badly in need of improvement.”, the customer mentions two aspects, i.e., “food” and “ambience”, and ex- presses positive sentiment towards the former and negative sentiment towards the latter. ∗Corresponding author Question-Answering (QA) Style Review - Question: Is [battery life] durable? How about [oper- ating speed] of the phone? - Answer: Yes, very durable but quite slow and obtuse. Aspect Sentiment Classification Towards QA - Input: QA text pair with given aspects - Output: [battery life]: Positive [operating speed]: Negative Figure 1: An example for illustrating the proposed task of Aspect Sentiment Classification towards Question- Answering (ASC-QA). Recently, a new interactive reviewing form, namely “Customer Question-Answering (QA)”, has become increasingly popular and a large-scale of such QA style reviews (as shown in Figure 1) could be found in several famous e-commerce platforms (e.g., Amazon and Taobao). Compared to traditional non-interactive customer reviews, interactive QA style reviews are more reliable and convincing because answer providers are ran- domly selected from the real customers who have purchased the product (Shen et al., 2018a). To well automatically-understand the QA style reviews, it’s worthwhile to perform ASC on the QA style reviews. However, we believe that Aspect Sentiment …

Content truncated to 500 words