Subtractive random forests with two choices

TL;DR

This paper extends Subtractive Random Forests to a multi-choice recommendation model, analyzing its effectiveness and robustness with diverse scenarios and user topic evolution, providing insights for practical multi-choice recommendation systems.

Contribution

Introduces a multi-choice recommendation model based on Subtractive Random Forests, analyzing its performance and robustness in various scenarios with user topic evolution.

Findings

Model performs well across diverse scenarios

Robustness to heavy-tailed time delays demonstrated

Insights into user topic evolution and system consistency

Abstract

Recommendation systems are pivotal in aiding users amid vast online content. Broutin, Devroye, Lugosi, and Oliveira proposed Subtractive Random Forests (\textsc{surf}), a model that emphasizes temporal user preferences. Expanding on \textsc{surf}, we introduce a model for a multi-choice recommendation system, enabling users to select from two independent suggestions based on past interactions. We evaluate its effectiveness and robustness across diverse scenarios, incorporating heavy-tailed distributions for time delays. By analyzing user topic evolution, we assess the system's consistency. Our study offers insights into the performance and potential enhancements of multi-choice recommendation systems in practical settings.