The paper segments e-commerce queries into broad and narrow types, uses autoencoders and embeddings to handle sparsity, and shows that separate pointwise or pairwise LETOR models for each type outperform a single combined model on fashion data.
Decoding fashion contexts using word embeddings
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2019 2verdicts
UNVERDICTED 2representative citing papers
A DNN combining BPR product embeddings and skip-gram body-shape embeddings predicts return probability before purchase in fashion e-commerce.
citing papers explorer
-
Learning to Rank Broad and Narrow Queries in E-Commerce
The paper segments e-commerce queries into broad and narrow types, uses autoencoders and embeddings to handle sparsity, and shows that separate pointwise or pairwise LETOR models for each type outperform a single combined model on fashion data.
-
Early Bird Catches the Worm: Predicting Returns Even Before Purchase in Fashion E-commerce
A DNN combining BPR product embeddings and skip-gram body-shape embeddings predicts return probability before purchase in fashion e-commerce.