A Transformer-Based Substitute Recommendation Model Incorporating Weakly Supervised Customer Behavior Data

TL;DR

This paper introduces a transformer-based substitute recommendation model that leverages product descriptions and weakly supervised customer data, improving accuracy by considering product functionality and supporting multiple languages.

Contribution

It proposes a novel language matching approach for substitute recommendation using product titles, with a de-noising transformation and multilingual support, deployed at scale.

Findings

Achieved a 19% revenue increase in online A/B testing.

Successfully deployed in 11 marketplaces across 6 languages.

Outperformed existing methods in offline and online evaluations.

Abstract

The substitute-based recommendation is widely used in E-commerce to provide better alternatives to customers. However, existing research typically uses the customer behavior signals like co-view and view-but-purchase-another to capture the substitute relationship. Despite its intuitive soundness, we find that such an approach might ignore the functionality and characteristics of products. In this paper, we adapt substitute recommendation into language matching problem by taking product title description as model input to consider product functionality. We design a new transformation method to de-noise the signals derived from production data. In addition, we consider multilingual support from the engineering point of view. Our proposed end-to-end transformer-based model achieves both successes from offline and online experiments. The proposed model has been deployed in a large-scale E-commerce website for 11 marketplaces in 6 languages. Our proposed model is demonstrated to increase revenue by 19% based on an online A/B experiment.