Integrating Predictive Models into Two-Sided Recommendations: A Matching-Theoretic Approach

TL;DR

This paper models two-sided recommendation as a matching problem, introducing a congestion-aware metric and an exposure-limited algorithm, which improves fairness and efficiency in dating platform recommendations through simulations and real-world experiments.

Contribution

It proposes a novel matching-theoretic framework with an exposure-constrained algorithm for two-sided recommendations, enhancing fairness and efficiency.

Findings

ECDA increases effective dates and receiver engagement.

Exposure control improves fairness without reducing overall engagement.

Field experiments validate the practical benefits of the proposed method.

Abstract

Two-sided platforms must recommend users to users, where matches (termed \emph{dates} in this paper) require mutual interest and activity on both sides. Naive ranking by predicted dating probabilities concentrates exposure on a small subset of highly responsive users, generating congestion and overstating efficiency. We model recommendation as a many-to-many matching problem and design integrators that map predicted login, like, and reciprocation probabilities into recommendations under attention constraints. We introduce \emph{effective dates}, a congestion-adjusted metric that discounts matches involving overloaded receivers. We then propose \emph{exposure-constrained deferred acceptance} (ECDA), which limits receiver exposure in terms of expected likes or dates rather than headcount. Using production-grade predictions from a large Japanese dating platform, we show in calibrated simulations that ECDA increases effective dates and receiver-side dating probability despite reducing total dates. A large-scale regional field experiment confirms these effects in practice, indicating that exposure control improves equity and early-stage matching efficiency without harming downstream engagement.