Recommending fashion outfits requires learning concepts of style and fashionability that are typically human.
In this paper, Federico Sallemi, Head of Product at ContentWise Showroom, and Gabriele Prato, Data Scientist at BCG, presented a novel machine learning algorithm for a fashion-specific recommender system, which they tested on data provided by fast-fashion global retailer H&M.
In this paper you will learn:
How the recommender system generates fashionable outfits, scales inference time, and more.
The preliminary results on how this algorithm can be employed in a real-world use case scenario.
The evaluations that three professional stylists provided on the outfits generated by our algorithm.
How we used the concept to develop ContentWise Showroom, our offering for fashion digital retail.