AutoInvent
Turn abandoned patents
into your own filing
Get Started
Patent drawing for Information extraction from natural language texts
US 10,691,891 B2
Computing US 10,691,891 B2 Not in force

Information extraction from natural language texts

This invention describes a system for extracting facts from written text. It works by using at least two different methods to pull out information, then specifically identifies when a first textual annotation associated with one extracted fact overlaps with a second textual annotation from another extracted fact. A special function then resolves these overlaps or conflicts, for instance, by modifying, deleting, or merging the conflicting information, to create a single, consistent list of facts.

Why it matters: Filed before the widespread adoption of large language models (LLMs), this invention's focus on arbitrating conflicting or overlapping information extractions is highly relevant today. LLMs often produce multiple interpretations or extractions, and combining their outputs with other techniques creates a greater need for robust conflict resolution mechanisms like those described.

Status
Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build
SpecializedSoftware logic, natural language processing algorithms

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.