Related Pins at Pinterest: The Evolution of a Real-World Recommender System

By David C. Liu, Stephanie Rogers, Raymond Shiau, Dmitry Kislyuk, Kevin C. K. Ma, Zhigang Zhong, Jenny Liu, Yushi Jing. 2017

139 citations recorded by the source corpus.

Related Pins is the Web-scale recommender system that powers over 40% of user engagement on Pinterest. This paper is a longitudinal study of three years of its development, exploring the evolution of the system and its components from prototypes to present state. Each component was originally built with many constraints on engineering effort and computational resources, so we prioritized the simplest and highest-leverage solutions. We show how organic growth led to a complex system and how we managed this complexity. Many challenges arose while building this system, such as avoiding feedback loops, evaluating performance, activating content, and eliminating legacy heuristics. Finally, we offer suggestions for tackling these challenges when engineering Web-scale recommender systems.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiangnan He
2. Tat-seng Chua
3. Xiang Wang
4. Julian McAuley
5. Zhankui He
6. Zhenguang Liu
7. Yongfeng Zhang
8. Jiaxi Tang

In Computer Vision and Pattern Recognition

1. Tat-seng Chua
2. Xiangnan He
3. Jingkuan Song
4. Dacheng Tao
5. Jure Leskovec
6. Ed H. Chi
7. Jiaxi Tang
8. Zhankui He

In Information Systems

1. Xiangnan He
2. Tat-seng Chua
3. Xiang Wang
4. Zhankui He
5. Zhenguang Liu
6. Jiaxi Tang
7. Ed H. Chi
8. Ke Wang

Citations in the corpus

• Pixie: A System for Recommending 3+ Billion Items to 200+ Million Users in Real-Time (2017, background): “Pixie is the primary source of candidates for Related Pins, though the full system features several additional layers [22].”
• Self-supervised Learning for Large-scale Item Recommendations (2020, background): “Deep learning has led to many successes in building industry-scale recommender systems such as video suggestion [10], news recommendation [34], and visual discovery in social networks [26, 40].”
• NAIS: Neural Attentive Item Similarity Model for Recommendation (2018, background): “Another common approach is to employ random walks on the user-item interaction graph [4].”
• Ranking Distillation: Learning Compact Ranking Models With High Performance for Recommender System (2018, methodology): “ds, such as pruning and indexing to speed-up retrieval of related items [23, 34], using fast models for candidate generation and applying time-consuming models to the candidates for online inferences [9, 24]. These methods either lose much of effectiveness, due to the introduced model constraints, o”