KGLit4Cat

GITHUB

SHORT DESCRIPTION

This tool enables automated extraction and organization of catalysis research data into a structured, FAIR-compliant knowledge graph. It combines ontology learning and named entity recognition (NER) to extract key information from scientific publications. The system builds on and improves the CatalysisIE model using a new dataset, enhancing precision and recall. The resulting knowledge graph supports user- and machine-readable formats, enabling easy access to relevant catalysis knowledge. Usability is demonstrated through validation on two catalysis datasets and example SPARQL queries for researchers.

PROBLEM STATEMENT

Catalysis research produces large volumes of unstructured data that are difficult to access, integrate, and reuse. The growing number of publications makes it challenging to efficiently extract relevant knowledge. The community needs automated, FAIR-aligned tools to transform literature into structured, searchable knowledge graphs that support data-driven research.

 

GITHUB