POG: Personalized Outfit Generation for Fashion Recommendation at Alibaba iFashion

TL;DR

This paper introduces POG, a personalized outfit generation model using Transformer architecture, which enhances fashion recommendation by ensuring compatibility and personalization, demonstrated through extensive experiments and deployment at Alibaba.

Contribution

The paper presents the first large-scale dataset of fashion outfits and user behaviors, and develops a Transformer-based model for personalized outfit generation in an industrial setting.

Findings

POG outperforms existing methods in compatibility and personalization metrics.

Extensive offline and online experiments validate the effectiveness of POG.

Successful deployment of POG on Alibaba's platform demonstrates practical utility.

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. 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 release a large-scale dataset consisting of 1.01 million outfits with rich context information, and 0.28 billion user click actions from 3.57 million users. To the best of our knowledge, this dataset is the largest, publicly available, fashion related dataset, and the first to provide user behaviors relating to both outfits and fashion items.