Back to all patents
Expiring soonaiAssigned to Amazon Technologies Inc

Recommendation system

Recommendation system using aggregated behavioral associations that maps items to related items for personalized shopping suggestions.

What it is

It analyzes users’ purchase and behavioral histories in aggregate to detect item-to-item associations, then stores them in an item-to-related-items mapping used for recommendations. The technical significance is that it turns patterns across many users into direct linkage scores, enabling fast “customers also viewed/bought” style retrieval without per-user model training. Previously, deploying this specific association-mapping approach for catalog recommendations could require licensing under the patent covering the recommendation mechanism.

What you could build

Launch e-commerce cross-sell and “related products” features powered by item-association mapping for marketplaces or subscription boxes.

Patent number
7908183B2
Expiration
Invalid Da
Assignee
Amazon Technologies Inc
Inventors
Jennifer A. Jacobi, Eric A. Benson, Gregory D. Linden

More in ai

Building on this patent?

Submit a problem you're solving with this technology and connect with founders, investors, and researchers.

Recommendation system | Expiring Patents | Questd | Questd