A hierarchical bayesian model for size recommendation in fashion

Guigourès, Romain; Ho, Yuen King; Koriagin, Evgenii; Sheikh, Abdul-Saboor; Bergmann, Urs; Shirvany, Reza

doi:10.1145/3240323.3240388

Cited by 29 publications

(44 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…2) We not only combine the advantages of the previous work in the field, but also overcome some of the limitations of the current state-of-the-art in size and fit recommendation. Unlike [5], our approach allows to use large amounts of data, deal with various size systems and exploit correlations across categories without any further expert knowledge (or manual mapping). In contrast to other deep-learning approaches [6,7], our approach represents customers by their purchase history so predictions can be updated very quickly and at low cost without retraining the model.…”

Section: The Contributions Of This Work Are 3-foldmentioning

confidence: 99%

“…Sembium et al [3] proposed a latent factor model and a follow-up Bayesian formulation [4] to predict a simple fit prediction (small, fit, large) given a size of a product for a customer. Other work recently proposed a hierarchical Bayesian model [5] for personalized size recommendation. Customers' purchase histories are used to model their size profile, and return reasons are leveraged to adjust size recommendations to specific products.…”

Section: Introductionmentioning

confidence: 99%

“…This model requires numerical sizes which does not readily apply to all fashion categories. [5] only considers size recommendation for shoes, extending it requires splitting gender and categories to align size mappings.…”

Section: Introductionmentioning

confidence: 99%

“…Following a deep learning formulation PSE [6] tackles the problem using a 2-step approach; embeddings for size recommendations are learnt first for (brand, size) pairs, then for (product, size) pairs. Due to the cardinality of the problem, one model is trained per gendercategory combination, as would be done in [5]. Such approaches involve a lot of manual work, but also greatly limit the quality of the abstractions learned.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Meta-Learning for Size and Fit Recommendation in Fashion

Lasserre¹,

Sheikh²,

Koriagin³

et al. 2020

Proceedings of the 2020 SIAM International Conference on Data Mining

Self Cite

View full text Add to dashboard Cite

Fashion e-commerce has enjoyed an exponential growth in the last few years. A key challenge of the market players is to offer customers a personalized experience and to suggest relevant articles. In that respect, although product recommendation is a well-studied field, size and fit recommendation is still in its infancy. The size and fit topic is a very challenging problem as data is extremely sparse and noisy. Most approaches so far have exploited traditional machine learning techniques. In this work, we bring forward a meta-learning approach using an underlying deep neural network. The advantage of such an approach lies in its ability to exploit large scale data, learn across fashion categories, and absorb new data efficiently without retraining. We benchmark our method against 3 recent methods proven successful in the domain, and demonstrate various strengths of the proposed approach. To that end, we use a large-scale anonymized dataset of about 9.4 million customer-size interactions, collected over 5 years from around 384k customers.

show abstract

Section: The Contributions Of This Work Are 3-foldmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Meta-Learning for Size and Fit Recommendation in Fashion

Lasserre¹,

Sheikh²,

Koriagin³

et al. 2020

Proceedings of the 2020 SIAM International Conference on Data Mining

Self Cite

View full text Add to dashboard Cite

show abstract

“…In [7], the authors propose a hierarchical Bayesian approach for personalized size recommendation. Conditioned on customer and article pairs, the method models the joint conditional probability of sizes ordered by customers together with their outcomes (i.e.…”

Section: Related Workmentioning

confidence: 99%

A deep learning system for predicting size and fit in fashion e-commerce

Sheikh¹,

Guigourès²,

Koriagin³

et al. 2019

Proceedings of the 13th ACM Conference on Recommender Systems

Self Cite

View full text Add to dashboard Cite

Personalized size and fit recommendations bear crucial significance for any fashion e-commerce platform. Predicting the correct fit drives customer satisfaction and benefits the business by reducing costs incurred due to size-related returns. Traditional collaborative filtering algorithms seek to model customer preferences based on their previous orders. A typical challenge for such methods stems from extreme sparsity of customer-article orders. To alleviate this problem, we propose a deep learning based content-collaborative methodology for personalized size and fit recommendation. Our proposed method can ingest arbitrary customer and article data and can model multiple individuals or intents behind a single account. The method optimizes a global set of parameters to learn population-level abstractions of size and fit relevant information from observed customer-article interactions. It further employs customer and article specific embedding variables to learn their properties. Together with learned entity embeddings, the method maps additional customer and article attributes into a latent space to derive personalized recommendations. Application of our method to two publicly available datasets demonstrate an improvement over the state-of-the-art published results. On two proprietary datasets, one containing fit feedback from fashion experts and the other involving customer purchases, we further outperform comparable methodologies, including a recent Bayesian approach for size recommendation.

show abstract