POG: Personalized Outfit Generation for Fashion Recommendation at Alibaba iFashion

TLDR

Increasing demand for fashion recommendation raises challenges for online shopping platforms, requiring outfit compatibility and personalization. The study aims to satisfy outfit compatibility and personalization by bridging outfit generation and recommendation. We propose a Transformer‑based Personalized Outfit Generation (POG) model that links user preferences for items and outfits, and deploy it on Alibaba’s Dida platform to generate personalized outfits for iFashion users. Data analysis shows users share similar tastes in items and outfits, and experiments demonstrate that POG outperforms alternatives on compatibility and personalization metrics.

Abstract

Increasing demand for fashion recommendation raises a lot of challenges for online shopping platforms and fashion communities. In particular, there exist two requirements for fashion outfit recommendation: the Compatibility of the generated fashion outfits, and the Personalization in the recommendation process. In this paper, we demonstrate these two requirements can be satisfied via building a bridge between outfit generation and recommendation. Through large data analysis, we observe that people have similar tastes in individual items and outfits. Therefore, we propose a Personalized Outfit Generation (POG) model, which connects user preferences regarding individual items and outfits with Transformer architecture. Extensive offline and online experiments provide strong quantitative evidence that our method outperforms alternative methods regarding both compatibility and personalization metrics. Furthermore, we deploy POG on a platform named Dida in Alibaba to generate personalized outfits for the users of the online application iFashion.