A Review of Modern Fashion Recommender Systems

Abstract

The textile and apparel industries have grown tremendously over the last years. Customers no longer have to visit many stores, stand in long queues, or try on garments in dressing rooms as millions of products are now available in online catalogs. However, given the plethora of options available, an effective recommendation system is necessary to properly sort, order, and communicate relevant product material or information to users. Effective fashion RS can have a noticeable impact on billions of customers' shopping experiences and increase sales and revenues on the provider-side.The goal of this survey is to provide a review of recommender systems that operate in the specific vertical domain of garment and fashion products. We have identified the most pressing challenges in fashion RS research and created a taxonomy that categorizes the literature according to the objective they are trying to accomplish (e.g., item or outfit recommendation, size recommendation, explainability, among others) and type of side-information (users, items, context). We have also identified the most important evaluation goals and perspectives (outfit generation, outfit recommendation, pairing recommendation, and fill-in-the-blank outfit compatibility prediction) and the most commonly used datasets and evaluation metrics. CCS Concepts: • Information systems → Recommender systems; Multimedia and multimodal retrieval; • Human-centered computing → User models.

Lightweight Model for Session-Based Recommender Systems with Seasonality Information in the Fashion Domain

“…It was also computed the feature importance of each attributes involved in the final model, in order to check its relevance. 5 The results show a balance between features, meaning that all of them contributed mostly equally in final ranking, with a slight preference for the scores of the baseline models. This suggests that the GBDTs model interpreted the scores as knowledgeable experts in specific contexts, selecting which ones to trust more according to the recommendation scenario of interest.…”

“…This is due to their usage of the 𝑈 𝑅𝑀, which also limits the models to take advantage of the temporal correlation between the session views. Additionally, it has to be taken into account that the fashion realm is a field in which the tastes of the customers are in continuous evolution, with different periods that often present far different tendencies on the purchased products [5]. This could lead the models, in some cases, to learn noisy patterns for the reference period.…”

The Unfairness of Active Users and Popularity Bias in Point-of-Interest Recommendation

“…Moreover, we observe that a primary challenge many models face is the unfairness of popularity bias. As for future work, we plan to extend our experiments on more datasets from different domains (e.g., marking in e-commerce [34], media-streaming, e-fashion [13]) and models such as sessionbased [28], neural [8] and content-based systems [15]. Investigating the different type of bias, such as gender bias, can be a potential direction for future research.…”

Reusable Self-Attention Recommender Systems in Fashion Industry Applications

“…The main objective of recommender systems is to recommend relevant content tailored to the customer's preferences and intent. Fashion e-commerce has heavily invested in developing recommender systems for different use cases to aid online shopping experience [4,8] including recommending relevant items [2,9,35], complete outfits [3,10,15], and size recommendation [14,23].…”