<p>Recent advances in machine learning have provided valuable tools for constructing various recommendation systems in e-commerce companies such as Amazon and eBay. In this paper, we analyze click history records from an online fashion mall using a well-known Bayesian topic model, the latent Dirichlet allocation (LDA). Although multinomial mixture models such as the LDA have been widely applied to analyze such count data, a basic LDA-based approach for fashion data analysis may yield a crucial issue. For a customer who clicked pants primarily, for example, a basic recommendation algorithm tends to recommend pants only. Given a click history of pants, a more desirable algorithm would recommend fashion items compatible with the clicked pants. For this purpose, we propose a novel Bayesian model called the group-constrained LDA, which can incorporate prior information about the item groups. The proposed method is applied to analyze the click history data from Samsung Fashion (SSF) Shop, one of the largest online fashion malls in South Korea.</p>