Exploiting LLMs for entity and schema extraction from unstructured documents (github.com)

I've been exploring ways to automate the extraction of structured data from unstructured documents like PDFs. This tool uses llm to identify relevant entities, relationships and generate schemas that can be used for database design or knowledge graph creation.

I'm curious about potential applications and improvements. Has anyone tackled similar challenges? What approaches have you found effective for handling unstructured data?