Ontology guided framework builds climate knowledge graph from papers
Ontology-Guided Knowledge Graph Construction from Structured Metadata and Climate Science Publications
Computation and LanguageArtificial Intelligence
Summary
Scientific papers grow so quickly that it’s hard for people to organize their knowledge manually. The authors developed a system that uses a large language model and structured rules from a climate science ontology to extract facts from papers and check them carefully. This system builds a large and reliable web of climate knowledge that links concepts and facts across thousands of studies. Their system matches human agreement levels in accuracy and can trace facts back to text in the original papers. They shared both the code and the resulting knowledge graph publicly.
What this means in practice
- •For environmental data engineers: Create a structured, validated knowledge graph from large sets of climate science publications for use in climate data analysis and decision support.
- •For scientific publishing platforms: Automate extraction and validation of factual knowledge from submitted climate research papers to improve metadata and discoverability.
Authors
Aayush Acharya, Qi Zhang, Eduard Dragut, Longin Jan Latecki
Abstract
The rapid expansion of scientific literature far outpaces the speed of manual curation, creating a fundamental bottleneck in extracting structured knowledge from scientific texts. To address this, we introduce an agentic, ontology-guided framework for knowledge graph (KG) construction. Our framework utilizes specialized agents for document understanding, information extraction, ontology conformance, and entity linking. First, an LLM extracts factual knowledge in the form of triples with explicit textual evidence from research papers. Then the triplets undergo a multistage verification process to ground the triplets in the ontology. By enforcing schema constraints and requiring evidence in the form of text fragments, we ensure the data is both semantically consistent and traceable. In an agreement study over 1,000 extracted triplets, Gemini-2.5-flash achieves Cohen's $κ$ scores of 0.640--0.700 against three human annotators, matching the human--human agreement range of 0.616--0.689. We applied this framework to 2,000 scientific publications and construct a novel, ontology-grounded KG for climate science with 133,990 nodes and 755,308 relations; of the 23,327 triplets that the publications contribute after validation, 98.96\% are judged correct. The pipeline code and KG is publicly available https://github.com/aayushacharya/climatekg.