Topic Driven Adaptive Network for Cross-Domain Sentiment Classification

TL;DR

This paper introduces TDAN, a novel transformer-based network that leverages topic models to extract domain-specific words, enhancing cross-domain sentiment classification by effectively utilizing domain-specific information.

Contribution

The paper proposes a new Topic Driven Adaptive Network that combines topic models with transformer-based attention mechanisms to improve cross-domain sentiment analysis.

Findings

TDAN outperforms existing methods in cross-domain sentiment classification.

Topic models help discover interpretable, low-dimensional subspaces.

Experimental results validate the effectiveness of the proposed approach.

Abstract

Cross-domain sentiment classification has been a hot spot these years, which aims to learn a reliable classifier using labeled data from a source domain and evaluate it on a target domain. In this vein, most approaches utilized domain adaptation that maps data from different domains into a common feature space. To further improve the model performance, several methods targeted to mine domain-specific information were proposed. However, most of them only utilized a limited part of domain-specific information. In this study, we first develop a method of extracting domain-specific words based on the topic information derived from topic models. Then, we propose a Topic Driven Adaptive Network (TDAN) for cross-domain sentiment classification. The network consists of two sub-networks: a semantics attention network and a domain-specific word attention network, the structures of which are based on transformers. These sub-networks take different forms of input and their outputs are fused as the feature vector. Experiments validate the effectiveness of our TDAN on sentiment classification across domains. Case studies also indicate that topic models have the potential to add value to cross-domain sentiment classification by discovering interpretable and low-dimensional subspaces.