NATUREKG: AN ONTOLOGY AND KNOWLEDGE GRAPH FOR NATURE FINANCE WITH A TEXT2CYPHER APPLICATION

This repository contains the resources associated with NatureKG, an ontology and knowledge graph developed for the domain of nature finance, with a focus on the built environment sector. Ontology schema & graph structure: Provided as a Neo4j dump file (neo4j.dump). This includes the ontology classes (e.g., Actions, Drivers of Nature Loss, Value Chains, Evidence, Sources) and their relationships (e.g., MITIGATES, ALIGNS_WITH, CITED_IN) instantiated in the NatureKG knowledge graph. Text2Cypher dataset: Provided as a JSON file (merged_cleaned_text_to_cypher.json). This dataset contains natural language question–query pairs designed for training and evaluating LLMs in translating text into Cypher queries over NatureKG. Built environment and source information: Provided as an Excel file (.xlsx) containing curated data sources and instantiation details from ENCORE, the Science Based Targets Network (SBTN), and scientific literature.