Online dating recommendations: matching markets and learning preferences

Kun Tu, Bruno F. Ribeiro, David D. Jensen, Don Towsley, Benyuan Liu, Hua Jiang, Xiaodong Wang

Abstract: Recommendation systems for online dating have recently attracted much attention from the research community. In this paper we propose a two-side matching framework for online dating recommendations and design an Latent Dirichlet Allocation (LDA) model to learn the user preferences from the observed user messaging behavior and user profile features. Experimental results using data from a large online dating website shows that two-sided matching improves the rate of successful matches by as much as 45%. Finally, using simulated matching, we show that the LDA model can correctly capture user preferences.