Inductive learning for product assortment graph completion

TL;DR

This paper introduces an inductive learning approach to improve product assortment graphs by leveraging textual and visual data, enhancing style compatibility predictions and benefiting transductive tasks in retail settings.

Contribution

It presents a novel inductive learning method that incorporates rich node information to enhance graph encoding of product compatibility in assortments.

Findings

Improved style compatibility prediction accuracy.

Enhanced performance on transductive graph tasks.

Minor impact on graph sparsity.

Abstract

Global retailers have assortments that contain hundreds of thousands of products that can be linked by several types of relationships like style compatibility, "bought together", "watched together", etc. Graphs are a natural representation for assortments, where products are nodes and relations are edges. Relations like style compatibility are often produced by a manual process and therefore do not cover uniformly the whole graph. We propose to use inductive learning to enhance a graph encoding style compatibility of a fashion assortment, leveraging rich node information comprising textual descriptions and visual data. Then, we show how the proposed graph enhancement improves substantially the performance on transductive tasks with a minor impact on graph sparsity.