This invention describes a method to understand relationships between different concepts, especially in life sciences. It works by identifying types for various words and phrases, then determining the type of a search term. The system then uses computer models called word embeddings to calculate how strongly these concepts are associated with each other. The claims narrow this to a method for generating these association scores.
Why it matters: Filed just as large language models were rapidly advancing. The underlying technology for generating sophisticated word embeddings and inferring complex semantic associations has become significantly more powerful and efficient since 2024.
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