e-Commerce product classification: our participation at cDiscount 2015 challenge

TL;DR

This paper details a team’s approach to the cDiscount 2015 challenge, achieving 64.20% accuracy in product classification using a weighted voting ensemble of linear models.

Contribution

The paper presents a successful text classification method with an ensemble of linear models for large-scale e-commerce product categorization.

Findings

Achieved 64.20% accuracy on private leaderboard

Ranked 10th out of 175 teams

Used a weighted voting ensemble of linear models

Abstract

This report describes our participation in the cDiscount 2015 challenge where the goal was to classify product items in a predefined taxonomy of products. Our best submission yielded an accuracy score of 64.20\% in the private part of the leaderboard and we were ranked 10th out of 175 participating teams. We followed a text classification approach employing mainly linear models. The final solution was a weighted voting system which combined a variety of trained models.