A Unified Framework of Semi-Supervised Neural Net with Application to eCommerce Image Classification

Background

The challenges faced by supervised learning for eCommerce image classification are well documented. Due to the requirement for enormous amount of training data, eCommerce use cases are often constrained by the inhibitive cost and time of data labeling. In the following training flow, is the step where we find ourselves spending up to 90% of our budget as well as time. In addition, wrong labels, as well as imbalances and biases in the labeled data, hamper the quality of supervised learning. …


Introduction

Image classification — the task of assigning an input image one label from a set of categories — is one of the classical tasks in supervised machine learning with a large variety of practical applications. In particular, image classification has been widely applied in eCommerce to extract product type. Whether the customer might be looking for a women’s shirt or a TV, product type serves a significant role throughout the entire online shopping experience. If we don’t get the product type right, it’s very difficult for the customer to navigate the catalog and quickly find the right product.

However, classifying…

Binwei Yang

Binwei is a Distinguished Data Scientist at Walmart Global Tech. His current interests span across computer vision and tooling for better ROI on data science.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store