Knowledge graph fusion

This invention describes a computer system that combines information from many different documents to create a single, larger knowledge base. It works by first identifying key facts and relationships within each document, creating a small "knowledge graph" for each one. Then, it looks for common facts or entities across these individual graphs and merges them together into one comprehensive graph, specifically using named-entity recognition on unstructured text and potentially a reading comprehension dataset like SQuAD.

Why it matters: Filed before the widespread adoption of powerful Large Language Models (LLMs). LLMs can now significantly simplify and improve the extraction of contextual information and the generation of knowledge graphs from unstructured text, making the initial steps of this invention much more robust and efficient.

Status: Not in forceListed as no longer active.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

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