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

TL;DR

This paper introduces the Brand-Topic Model (BTM), an adversarially trained Poisson factorisation approach that automatically infers continuous sentiment scores and fine-grained sentiment-topics from product reviews, improving brand sentiment analysis.

Contribution

The paper presents a novel adversarial learning framework for Poisson factorisation that captures continuous sentiment variations and generates detailed sentiment-topics, surpassing existing models.

Findings

BTM outperforms baselines in brand ranking accuracy.

BTM achieves better topic coherence and separation.

BTM effectively models continuous sentiment changes.

Abstract

In this paper, we propose the Brand-Topic Model (BTM) which aims to detect brand-associated polarity-bearing topics from product 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 gradually varies from negative to positive. BTM is built on the Poisson factorisation model with the incorporation of adversarial learning. It has been evaluated on a dataset constructed from Amazon reviews. Experimental results show that BTM outperforms a number of competitive baselines in brand ranking, achieving a better balance of topic coherence and uniqueness, and extracting better-separated polarity-bearing topics.