Time-sensitive recommendation from recurrent user activities

TLDR

Recommender systems enhance user engagement, yet most algorithms ignore temporal behavior and recurrent user activities, leaving open how to recommend the right item at the right moment and predict when a user will return. The authors propose a framework linking self‑exciting point processes with low‑rank models to capture recurrent temporal patterns in user‑item interactions. The framework models user‑item consumption as a self‑exciting point process with a low‑rank structure, enabling efficient estimation via convex optimization. The model’s parameters are estimated by convex optimization, and an O(1/ε) algorithm scales to millions of pairs and hundreds of millions of events, yielding superior predictive performance on synthetic and real datasets and allowing integration of additional user context.

Abstract

By making personalized suggestions, a recommender system is playing a crucial role in improving the engagement of users in modern web-services. However, most recommendation algorithms do not explicitly take into account the temporal behavior and the recurrent activities of users. Two central but less explored questions are how to recommend the most desirable item at the right moment, and how to predict the next returning time of a user to a service. To address these questions, we propose a novel framework which connects self-exciting point processes and low-rank models to capture the recurrent temporal patterns in a large collection of user-item consumption pairs. We show that the parameters of the model can be estimated via a convex optimization, and furthermore, we develop an efficient algorithm that maintains O(1/∊) convergence rate, scales up to problems with millions of user-item pairs and hundreds of millions of temporal events. Compared to other state-of-the-arts in both synthetic and real datasets, our model achieves superb predictive performance in the two time-sensitive recommendation tasks. Finally, we point out that our formulation can incorporate other extra context information of users, such as profile, textual and spatial features.