Churn-Aware Recommendation Planning under Aggregated Preference Feedback

TL;DR

This paper introduces a privacy-aware recommendation planning model that infers user preferences from aggregated feedback, balancing exploration and exploitation to reduce user churn and improve personalization.

Contribution

It proposes the Rec-APC model for sequential decision-making with aggregated preference feedback, including a Bayesian approach and an efficient algorithm for optimal policies.

Findings

Rapid convergence to effective recommendations in experiments.

Outperforms existing POMDP solver SARSOP in large user type scenarios.

Applicable to real-world datasets like MovieLens.

Abstract

We study a sequential decision-making problem motivated by recent regulatory and technological shifts that limit access to individual user data in recommender systems (RSs), leaving only population-level preference information. This privacy-aware setting poses fundamental challenges in planning under uncertainty: Effective personalization requires exploration to infer user preferences, yet unsatisfactory recommendations risk immediate user churn. To address this, we introduce the Rec-APC model, in which an anonymous user is drawn from a known prior over latent user types (e.g., personas or clusters), and the decision-maker sequentially selects items to recommend. Feedback is binary -- positive responses refine the posterior via Bayesian updates, while negative responses result in the termination of the session. We prove that optimal policies converge to pure exploitation in finite time and propose a branch-and-bound algorithm to efficiently compute them. Experiments on synthetic and MovieLens data confirm rapid convergence and demonstrate that our method outperforms the POMDP solver SARSOP, particularly when the number of user types is large or comparable to the number of content categories. Our results highlight the applicability of this approach and inspire new ways to improve decision-making under the constraints imposed by aggregated preference data.