An Embedding-based Joint Sentiment-Topic Model for Short Texts

TL;DR

This paper introduces ELJST, an embedding-enhanced joint sentiment-topic model that improves the coherence and diversity of topics extracted from short texts, aiding better understanding of user behavior.

Contribution

The paper presents a novel embedding-based generative model with a Markov Random Field regularizer for more coherent and diverse short text topics, leveraging higher-order semantic information.

Findings

10% improvement in topic coherence

5% increase in topic diversification

Enhanced understanding of user behavior

Abstract

Short text is a popular avenue of sharing feedback, opinions and reviews on social media, e-commerce platforms, etc. Many companies need to extract meaningful information (which may include thematic content as well as semantic polarity) out of such short texts to understand users' behaviour. However, obtaining high quality sentiment-associated and human interpretable themes still remains a challenge for short texts. In this paper we develop ELJST, an embedding enhanced generative joint sentiment-topic model that can discover more coherent and diverse topics from short texts. It uses Markov Random Field Regularizer that can be seen as a generalisation of skip-gram based models. Further, it can leverage higher-order semantic information appearing in word embedding, such as self-attention weights in graphical models. Our results show an average improvement of 10% in topic coherence and 5% in topic diversification over baselines. Finally, ELJST helps understand users' behaviour at more granular levels which can be explained. All these can bring significant values to the service and healthcare industries often dealing with customers.