Modeling Churn in Recommender Systems with Aggregated Preferences

TL;DR

This paper introduces a model for recommender systems that uses aggregated user preferences to balance exploration and exploitation, aiming to reduce user churn under data privacy constraints.

Contribution

It proposes a probabilistic model with optimal policies that transition from exploration to exploitation, along with a branch-and-bound algorithm for policy computation.

Findings

Optimal policies transition from exploration to exploitation over time

The branch-and-bound algorithm effectively computes these policies

Empirical validation shows improved recommendation stability

Abstract

While recommender systems (RSs) traditionally rely on extensive individual user data, regulatory and technological shifts necessitate reliance on aggregated user information. This shift significantly impacts the recommendation process, requiring RSs to engage in intensive exploration to identify user preferences. However, this approach risks user churn due to potentially unsatisfactory recommendations. In this paper, we propose a model that addresses the dual challenges of leveraging aggregated user information and mitigating churn risk. Our model assumes that the RS operates with a probabilistic prior over user types and aggregated satisfaction levels for various content types. We demonstrate that optimal policies naturally transition from exploration to exploitation in finite time, develop a branch-and-bound algorithm for computing these policies, and empirically validate its effectiveness.