Employing hierarchical Bayesian networks in simple and complex emotion topic analysis


Traditional emotion models, when tagging single emotions in documents, often ignore the fact that most documents convey omplex human emotions. In this paper, we join emotion analysis with topic models to find complex emotions in documents, as ell as the intensity of the emotions, and study how the document emotions vary with topics. Hierarchical Bayesian networks re employed to generate the latent topic variables and emotion variables. On average, our model on single emotion classification utperforms the traditional supervised machine learning models such as SVM and Naive Bayes. The other model on the complex motion classification also achieves promising results. We thoroughly analyze the impact of vocabulary quality and topic quantity o emotion and intensity prediction in our experiments. The distribution of topics such as Friend and Job are found to be sensitive to he documents’ emotions, which we call emotion topic variation in this paper. This reveals the deeper relationship between topics nd emotions. 2012 Elsevier Ltd. All rights reserved.

DOI: 10.1016/j.csl.2012.07.012

14 Figures and Tables

Cite this paper

@article{Ren2013EmployingHB, title={Employing hierarchical Bayesian networks in simple and complex emotion topic analysis}, author={Fuji Ren and Xin Kang}, journal={Computer Speech & Language}, year={2013}, volume={27}, pages={943-968} }