Artificial intelligence system for balancing relevance and diversity of network-accessible content

This invention describes an AI system that recommends content by balancing how relevant and how diverse the recommendations are. It first trains a machine learning model, like a neural network, to score how relevant items from one catalog are to a user, based on their past interactions with a different catalog. Then, it uses these relevance scores along with a diversity goal to select a final set of recommendations that are both highly relevant and varied.

Why it matters: Filed before the widespread availability of highly optimized machine learning frameworks and cloud-based AI infrastructure. Generating sophisticated item vector representations (claim 3) and training complex neural networks (claim 2) for cross-catalog recommendations (claim 1) is now significantly more efficient and accessible.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedMachine learning, software logic, neural networks

3 AI reinvention directions for this patent

See all three

AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.

Reinvent this with AI

Where this one came from

Every morning we go through millions of abandoned patents and pick out 15 worth your time. This was one of today’s.

This one

+10