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

TL;DR

This paper introduces effective dates, a congestion-adjusted metric that discounts matches involving overloaded receivers and proposes exposure-constrained deferred acceptance (ECDA), which limits receiver exposure in terms of expected likes or dates rather than headcount.

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.

Figures (20)

• Figure 1: Chart Illustrating the Process from Recommendation to Date Formation
• Figure 2: ROC Curves
• Figure 3: Calibration Curves
• Figure 5: Likes, One-sided and DA
• Figure 6: Likes, ECDA

Theorems & Definitions (4)

• Theorem 1
• Theorem 2
• proof
• proof