How to Grow a (Product) Tree: Personalized Category Suggestions for eCommerce Type-Ahead

TL;DR

This paper introduces SessionPath, a neural network model that enhances personalized category suggestions in eCommerce search by leveraging session embeddings and explicit taxonomy path predictions, improving relevance and scalability.

Contribution

The paper presents SessionPath, a novel neural network that combines session-based personalization with explicit taxonomy path prediction for better facet suggestions.

Findings

Outperforms count-based and neural models in benchmarks

Enables scalable personalization in eCommerce search

Balances business needs with model behavior

Abstract

In an attempt to balance precision and recall in the search page, leading digital shops have been effectively nudging users into select category facets as early as in the type-ahead suggestions. In this work, we present SessionPath, a novel neural network model that improves facet suggestions on two counts: first, the model is able to leverage session embeddings to provide scalable personalization; second, SessionPath predicts facets by explicitly producing a probability distribution at each node in the taxonomy path. We benchmark SessionPath on two partnering shops against count-based and neural models, and show how business requirements and model behavior can be combined in a principled way.