KDD Papers

Unsupervised P2P Rental Recommendations via Integer Programming

Yanjie Fu (Missouri University of Science and Technology);Guannan Liu (Beihang University);Mingfei Teng (Rutgers University);Charu Aggarwal (IBM T. J. Watson Research Center)


Due to the sparseness of quality rating data, unsupervised recommender systems are used in many applications in Peer to Peer (P2P) rental marketplaces such as Airbnb, FlipKey, and HomeAway. We present an integer programming based recommender systems, where both accommodation benefit and community risk of lodging places are measured and are incorporated into objective function as utility measurements. More specifically, we first present an unsupervised fused scoring method for quantifying the accommodation benefit and community risk of a lodging with crowd-sourced geo-tagged data. In the view of maximizing the utility of recommendations, we formulate the unsupervised P2P rental recommendations as a constrained integer programming problem, where the accommodation benefit of recommendations is maximized and the community risk of recommendations is minimized, while maintaining constraints on personalization. Furthermore, we provide an e fficient solution for the optimization problem by developing a learning to integer programming method for combining aggregated listwise learning to rank into branching variable selection. We apply the proposed approach to the Airbnb data of New York City and provide lodging recommendations to travelers. In empirical experiments, we demonstrate the effectiveness of our method in striking a trade-off among satisfaction time on market, number of reviews, and achieving a balance between positive and negative sides, as well as the effi ciency enhancement of our methods.