This invention describes a computer system for building a structured knowledge base, called an ontology. It works by defining a fixed set of concept types and relationship types, then creating facts by linking two concepts with a relationship. The system can also track where each piece of information came from, including specific documents or data sources.
Why it matters: Filed before advanced AI could automatically extract concepts and relationships from text. Large Language Models now make populating such a structured knowledge base from documents, as described in the claims, significantly more efficient and scalable.
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