PinnerSage: Multi-Modal User Embedding Framework for Recommendations at Pinterest

TL;DR

PinnerSage is a multi-modal user embedding framework for Pinterest that improves personalized recommendations by capturing diverse user interests through hierarchical clustering and representative pins, outperforming single embedding methods.

Contribution

Introduces PinnerSage, a novel multi-modal user embedding system using hierarchical clustering and representative pins for enhanced recommendation quality.

Findings

Significantly outperforms single embedding methods in offline and online tests.

Successfully deployed at Pinterest at large scale.

Provides interpretable and efficient user representations.

Abstract

Latent user representations are widely adopted in the tech industry for powering personalized recommender systems. Most prior work infers a single high dimensional embedding to represent a user, which is a good starting point but falls short in delivering a full understanding of the user's interests. In this work, we introduce PinnerSage, an end-to-end recommender system that represents each user via multi-modal embeddings and leverages this rich representation of users to provides high quality personalized recommendations. PinnerSage achieves this by clustering users' actions into conceptually coherent clusters with the help of a hierarchical clustering method (Ward) and summarizes the clusters via representative pins (Medoids) for efficiency and interpretability. PinnerSage is deployed in production at Pinterest and we outline the several design decisions that makes it run seamlessly at a very large scale. We conduct several offline and online A/B experiments to show that our method significantly outperforms single embedding methods.