Revolutionizing Fashion Recommendations: A Deep Dive into Deep Learning-based Recommender Systems
Ilham Kachbal, Saîd El Abdellaoui, Khadija Arhid · 2024
Fashion takes center stage in our exploration of cutting-edge recommender systems. This paper presents an innovative approach to recommender systems for fashion that divides them into two key components: context-aware recommendations and outfit-based recommendations. Through the use of deep learning techniques, we explore personalized fashion discovery by considering contextual factors like climate and occasion to provide more relevant suggestions. At the same time, our research pioneers the approach of outfit-based recommendations that go beyond individual items by taking into account how complete outfits work together. This new way of thinking about recommender systems unveils a method where each suggested outfit is personalized by taking into account contextual influences as well as synergies between items, resulting in recommendations that could transform the future of customized fashion experiences for users.