2017 IEEE 42nd Conference on Local Computer Networks (LCN) (2017)
Oct. 9, 2017 to Oct. 12, 2017
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/LCN.2017.24
As a shared economy platform, Airbnb provides collaborative practices for customers and guides them to match with hosts' rooms. Based on the records and ratings, there is great significance attached to inferring the satisfaction between users and rooms. Several essential problems arise when evaluating satisfaction and matching. Data confidence and prediction bias influence the inference performance of the satisfaction. When two users stay in a room, the two users' joint satisfaction also deserves particular research because of the roommate effect. In this paper, the satisfaction is inferred considering confidence and prediction uncertainties. The satisfaction with the confidence uncertainty is modeled using a normalized variance of the Beta distribution. The algorithms for inferring satisfaction with the prediction uncertainties are divided into two parts: a weighted matrix factorization-based algorithm for individuals and a preference similarity-based algorithm for pairs. The problem can be reduced to a matching problem. Finally, extensive experiments show the effectiveness and accuracy of the proposed method.
customer satisfaction, matrix decomposition, recommender systems, service industries
L. Guo, J. Wu, W. Chang, J. Wu and J. Li, "Proposed Matching Scheme with Confidence and Prediction Uncertainty in Shared Economy," 2017 IEEE 42nd Conference on Local Computer Networks (LCN), Singapore, Singapore, 2018, pp. 591-594.