KGGen: Extracting Knowledge Graphs from Plain Text with Language Models
4 months ago
- #natural-language-processing
- #machine-learning
- #knowledge-graphs
- KGGen is a tool that extracts knowledge graphs from plain text using language models.
- It addresses the scarcity of high-quality knowledge graphs by clustering related entities to reduce sparsity.
- KGGen is available as a Python library, making it accessible for widespread use.
- A new benchmark called MINE (Measure of Information in Nodes and Edges) is introduced to evaluate KG extractors.
- KGGen demonstrates superior performance compared to existing KG extraction tools.