Domain adaptation has been popularly studied on exploiting labeled information from a source domain to learn a prediction model in a target domain. In this paper, we develop a novel representation learning approach to address domain adaptation for text classification with automatically induced discriminative latent features, which are generalizable across domains while informative to the prediction task. Specifically, we propose a hierarchical multinomial Naive Bayes model with latent variables to conduct supervised word clustering on labeled documents from both source and target domains, and then use the produced cluster distribution of each word as its latent feature representation for domain adaptation. We train this latent graphical model using a simple expectation-maximization (EM) algorithm. We empirically evaluate the proposed method with both cross-domain document categorization tasks on Reuters-21578 dataset and cross-domain sentiment classification tasks on Amazon product review dataset. The experimental results demonstrate that our proposed approach achieves superior performance compared with alternative methods.

2013 Conference on Empirical Methods in Natural Language Processing, EMNLP 2013
School of Computer Science

Xiao, M. (Min), Zhao, F. (Feipeng), & Guo, Y. (2013). Learning latent word representations for domain adaptation using supervised word clustering. In EMNLP 2013 - 2013 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference (pp. 152–162).