HKUST-KnowComp/AutoSchemaKG

This repository contains the implementation of AutoSchemaKG, a novel framework for automatic knowledge graph construction that combines schema generation via conceptualization.

55
/ 100
Established

Employs a two-stage LLM-based pipeline that extracts entity/event triples from unstructured text, then automatically induces schemas through conceptualization to enable zero-shot cross-domain inference. Supports batch processing with configurable LLM backends (OpenAI, Hugging Face), vector storage integration, and Neo4j hosting for large-scale graphs. Pre-built ATLAS knowledge graphs contain 900M+ nodes across Wikipedia, academic papers, and web content, with evaluation frameworks for factuality (FELM) and general performance (MMLU).

707 stars.

No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 15 / 25
Community 20 / 25

How are scores calculated?

Stars

707

Forks

90

Language

Python

License

MIT

Last pushed

Jan 14, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/HKUST-KnowComp/AutoSchemaKG"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.