POG: Personalized Outfit Generation for Fashion Recommendation at Alibaba iFashion

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 how these two requirements can be satisfied via building a bridge between outfit generation and recommendation. Through large data analysis, we find that people have similar tastes in individual items and outfits. Therefore, we propose a Personalized Outfit Generation model named POG, which connects user preferences regarding individual items and outfits with Transformer architecture. Multiple offline and online experiments provide strong quantitative evidence that our framework 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. This work represents a first step towards an industrial-scale fashion outfit generation and recommendation solution, which goes beyond generating outfits based on explicit queries, or merely recommending from existing outfit pools. As part of this work, we have collected a large-scale dataset consisting of 1.05 million outfits with rich context information, 1.05 billion user click actions from 5.42 million users. To the best our knowledge, our dataset is the largest, publicly available, fashion related dataset, and the first to provide user behaviors relating to both outfits and fashion items.