Anaphora resolution for medical text with machine learning and relevance …

This invention describes a computer program that figures out what a pronoun or similar word (an anaphor) refers to in medical text. It works by identifying all the important people, places, or things (entities) and the anaphor in a sentence. Then, it uses existing medical knowledge to find connections between these entities. It creates special data points (tuples) that link the anaphor to the thing it refers to (antecedent) and another related entity, using these connections to ultimately decide which entity the anaphor is referencing.

Why it matters: Filed before large language models (LLMs) became prevalent, this invention's reliance on explicit domain knowledge and structured tuple construction for anaphora resolution could now be significantly enhanced or simplified by LLMs' inherent contextual understanding and ability to implicitly leverage vast knowledge bases.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedSoftware logic, machine learning, domain knowledge

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