Skip to content
Product categorization using sequence-to-sequence models and others.
Jupyter Notebook Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
created_category
tutorial
LICENSE Initial commit Jul 3, 2018
README.md Update README.md Jul 10, 2018
Seq2Seq Product Categorization.ipynb Minor wording change Jul 5, 2018
example.attention Add attention visualisation Jul 4, 2018
neko-poster-final.pdf Added poster Jul 10, 2018
unconstrained-product-categorization-seq2seq.pdf

README.md

neko

This is a supplementary material for our publication to SIGIR ECOM'18 workshop, where we redefine the classic product categorization task as a sequence generation task.

This repository contains the following:

Abstract

Product categorization is a critical component of e-commerce plat- forms that enables organization and retrieval of the relevant products. Instead of following the conventional classification approaches, we consider category prediction as a sequence generation task where we allow product categorization beyond the hierarchical definition of the full taxonomy. Our paper presents our submissions for the Rakuten Data Challenge at SIGIR eCom’18. The goal of the challenge is to predict the multi-level hierarchical product categories given the e-commerce product titles. We ensembled several attentional sequence-to-sequence models to generate product category labels without supervised constraints. Such unconstrained product categorization suggests possible addition to the existing category hierarchy and reveals ambiguous and repetitive category leaves.

Our system achieved a balanced F-score of 0.8256, while the organizers’ baseline system scored 0.8142, and the best performing system scored 0.8513.

Cite

Yundi Maggie Li, Liling Tan, Stanley Kok and Ewa Szymanska. 2018. Unconstrained Production Categorization with Sequence-to-Sequence Models. In Proceedings of SIGIR 2018 Workshop on eCommerce (ECOM 18).

Bibtex:

@inproceedings{maggie18,
author= {Yundi Maggie Li and Liling Tan and Stanley Kok and Ewa Szymanska},
title={Unconstrained Product Categorization with Sequence-to-Sequence Models},
booktitle={{SIGIR 2018 Workshop on eCommerce (ECOM 18)}},
year=2018,
}
You can’t perform that action at this time.