Prediction and Congestion in Two-Sided Markets: Economist versus Machine Matchmakers
Posted: 31 Jan 2019 Last revised: 8 Apr 2019
Date Written: March 3, 2019
We study the recommender systems on the two-sided platforms. We find that the machine-learning algorithms create congestion: they often generate recommendations that are concentrated on a few users. We propose equilibrium machine-learning algorithms: they inherit the predicting power from machine-learning and solve the congestion problem by the market allocation mechanism in economics. We apply our recommenders to an online dating service that contains over 490,000 unique users. The hit rate of our equilibrium recommender outperforms the baseline content filtering by a factor of 19. In the counterfactual simulations, it accelerates the matching process by 200%.
Keywords: Online Dating, Two-Sided Matching, Recommender Systems, Matrix Factorization, Machine Learning, Reciprocal Recommender, Collaborative Filtering
JEL Classification: C53, C78, L81
Suggested Citation: Suggested Citation