Taxonomy-driven computation of product recommendations

TLDR

Recommender systems have surged in popularity, and large product taxonomies—such as Amazon’s hand‑crafted classifications—are increasingly common, enabling detailed machine‑readable content descriptions. This work exploits such taxonomic knowledge to compute personalized product recommendations. The approach leverages super‑concept/sub‑concept relationships to infer user profiles from the classifications of products they have chosen. Offline and online experiments show the method outperforms standard baselines when user data is sparse and only implicit ratings are available.

Abstract

Recommender systems have been subject to an enormous rise in popularity and research interest over the last ten years. At the same time, very large taxonomies for product classification are becoming increasingly prominent among e-commerce systems for diverse domains, rendering detailed machine-readable content descriptions feasible. Amazon.com makes use of an entire plethora of hand-crafted taxonomies classifying books, movies, apparel, and various other goods. We exploit such taxonomic background knowledge for the computation of personalized recommendations. Hereby, relationships between super-concepts and sub-concepts constitute an important cornerstone of our novel approach, providing powerful inference opportunities for profile generation based upon the classification of products that customers have chosen. Ample empirical analysis, both offline and online, demonstrates our proposal's superiority over common existing approaches when user information is sparse and implicit ratings prevail.