Adversarial Learning of Poisson Factorisation Model for Gauging Brand Sentiment in User Reviews

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, pages 2341–2351 April 19 - 23, 2021. ©2021 Association for Computational Linguistics 2341 Adversarial Learning of Poisson Factorisation Model for Gauging Brand Sentiment in User Reviews Runcong Zhao, Lin Gui, Gabriele Pergola, Yulan He Department of Computer Science, University of Warwick, UK {runcong.zhao,lin.gui,gabriele.pergola,yulan.he}@warwick.ac.uk Abstract In this paper, we propose the Brand-Topic Model (BTM) which aims to detect brand- associated polarity-bearing topics from prod- uct reviews. Different from existing models for sentiment-topic extraction which assume topics are grouped under discrete sentiment categories such as ‘positive’, ‘negative’ and ‘neural’, BTM is able to automatically infer real-valued brand-associated sentiment scores and generate fine-grained sentiment-topics in which we can observe continuous changes of words under a certain topic (e.g., ‘shaver’ or ‘cream’) while its associated sentiment gradu- ally varies from negative to positive. BTM is built on the Poisson factorisation model with the incorporation of adversarial learning. has been evaluated on a dataset constructedIt from Amazon reviews. Experimental results show that BTM outperforms a number of com- petitive baselines in brand ranking, achieving a better balance of topic coherence and unique- ness, and extracting better-separated polarity- bearing topics. 1 Introduction Market intelligence aims to gather data from a com- pany’s external environment, such as customer sur- veys, news outlets and social media sites, in or- der to understand customer feedback to their prod- ucts and services and to their competitors, for a better decision making of their marketing strate- gies. Since consumer purchase decisions are heav- ily influenced by online reviews, it is important to automatically analyse customer reviews for on- line brand monitoring. Existing sentiment analysis models either classify reviews into discrete polarity categories such as ‘positive’, ‘negative’ or ‘neural’, or perform more fine-grained sentiment analysis, in which aspect-level sentiment label is predicted, though still in the discrete polarity category space. We argue that it is desirable to be able to detect subtle topic changes under continuous sentiment scores. This allows us to identify, for example, whether customers with slightly negative views share similar concerns with those holding strong negative opinions; and what positive aspects are praised by customers the most. In addition, deriv- ing brand-associated sentiment scores in a continu- ous space makes it easier to generate a ranked list of brands, allowing for easy comparison. Existing studies on brand topic detection were largely built on the Latent Dirichlet Allocation (LDA) model (Blei et al., 2003) which assumes that latent topics are shared among competing brands for a certain market. They however are not able to separate positive topics from negative ones. Ap- proaches to polarity-bearing topic detection can only identify topics under discrete polarity cate- gories such as ‘positive’ and ‘negative’. We instead assume that each brand is associated with a latent real-valued sentiment score falling into the range of [−1, 1] in which −1 denotes negative, 0 being neutral and 1 positive, and propose a Brand-Topic Model built on the Poisson Factorisation model with adversarial …

Content truncated to 500 words