Cross-Domain Review Generation for Aspect-Based Sentiment Analysis

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 4767–4777 August 1–6, 2021. ©2021 Association for Computational Linguistics 4767 Cross-Domain Review Generation for Aspect-Based Sentiment Analysis Jianfei Yu, Chenggong Gong, and Rui Xia∗ School of Computer Science and Engineering, Nanjing University of Science and Technology, China {jfyu, cggong, rxia}@njust.edu.cn Abstract Supervised learning methods have proven to be effective for Aspect-Based Sentiment Anal- ysis (ABSA). However, the lack of fine- grained labeled data hinders their effective- ness in many domains. To address this is- sue, unsupervised domain adaptation methods are desired to transfer knowledge from a la- beled source domain to any unlabeled target domain. In this paper, we propose a new do- main adaptation paradigm called cross-domain review generation (CDRG), which aims to gen- erate target-domain reviews with fine-grained annotation based on the source-domain la- beled reviews. To achieve this goal, we pro- pose a two-step approach as a concrete real- ization of CDRG. It first converts a source- domain review to a domain-independent re- view by masking its source-specific attributes, and then converts the domain-independent re- view to a target-domain review with a masked language model pre-trained in the target do- main. We further propose two ways to lever- age the generated target-domain reviews for two cross-domain ABSA tasks. Extensive ex- periments demonstrate the superiority of our CDRG-based approaches over the state-of-the- art domain adaptation methods. 1 Introduction Aspect-Based Sentiment Analysis (ABSA) has re- ceived considerable attention in recent years (Liu, 2020). The goal of ABSA is to extract the aspect terms mentioned in review sentences and predict the sentiments over them (Pontiki et al., 2016). For example, given the review “The fish soup is deli- cious”, the aspect term and its sentiment are fish soup and Positive, respectively. With the recent advances in deep learning, many supervised neural models have been proposed for several ABSA tasks, e.g., aspect extraction (Liu et al., 2015; Xu et al., ∗Corresponding author. 2018), aspect-level sentiment classification (Wang et al., 2016b; Tang et al., 2016), and End-to-End ABSA (Zhang et al., 2015; Li et al., 2019a). Although these neural models have obtained promising results in several product domains such as Laptop and Restaurant (Pontiki et al., 2014), the major obstacle to them is the lack of rich an- notated resources in many new domains. Since ABSA requires fine-grained annotation of aspect terms and their sentiments in each review, it is ex- tremely time-consuming to develop such resources for each domain, and the annotation process can be prohibitively expensive. To alleviate the annotation efforts, unsupervised domain adaptation methods are desired to transfer knowledge from a source domain with rich labeled data to a target domain with only unlabeled data. The key challenge of domain adaptation is that the data distribution of the source domain usually differs from that of the target domain. To allevi- ate the domain discrepancy, many approaches have been proposed in coarse-grained sentiment clas- sification based on the following two paradigms: • feature-based adaptation, which aims to aims to learn …

Content truncated to 500 words