Qiu, Minghui, et al. "Modeling user arguments, interactions, and attributes for stance prediction in online debate forums." Proceedings of the 2015 SIAM international conference on data mining. Society for Industrial and Applied Mathematics, 2015.

Online debate forums are key platforms for expressing and discussing opinions, but often suffer from limited user participation on specific issues, which leads to sparse stance data. This study addresses the challenge of stance prediction by leveraging additional information available in forums, such as user arguments, interactions, and biographical details. An integrated model is proposed that uses a regression-based latent factor approach to jointly analyze these types of side information. The model enhances stance prediction for both users with prior engagement (warm-start) and new users (cold-start). Experimental results show that this method performs well in predicting stances at both micro and macro levels.