Climate Knowledge Graph has structured the Intergovernmental Panel on Climate Change’s (IPCC) Sixth Assessment Report (AR6) as a Wikibase knowledge graph.

The knowledge graph is made of five datasets:

1. Text corpus: Structured hierarchy of AR6 reports, chapters, divisions, and full text.
2. Bibliography: Bibliographic information for reports and chapters.
3. Glossary: Terms and definitions for climate concepts.
4. Acronyms: Acronym index with expanded forms and references.
5. Authors: Author identities and links to reports.

An invitation to data scientists

The knowledge graph acts as a structural (syntactic) framework for data scientists to add Linked Open Data (LOD) about the scientific findings contained in the report.

Add to the knowledge graph or make use of the data yourself.

Find out more

How the knowledge graph is made

Web to Wikibase: ClimateKG has made a software pipeline to structure text corpora like the IPCC AR6 report.

The AR6 report is web scraped from the IPCC AR6 website and follows a CSV → XML/DTD → Wikibase import pipeline. An Entity-Relationship Model (ER Model) maps every DTD element to a confirmed Wikibase property.

NotePipeline Formula

Web scrape + ER model = knowledge graph

The steps—tools and resources:

Tools & Resources	Description
CKG Scrape	Web scraping workflow used to extract and structure the IPCC AR6 report.
Datasets	Corpus · Authors · Acronyms · Bibliography · Glossary
ER Model	Interactive entity-relationship diagram — fullscreen, zoomable, and showing the corpus hierarchy: Work (Q2) → Report Series (Q3) → Report (Q4) → Text Division (Q5) → Chapter (Q6)
XML/DTD	Validated XML and DTD layer for structuring data before import into Wikibase, with XSLT transformation for reusable outputs.
Knowledge Graph	Wikibase for structured data and linked relationships across the report corpus.
Browse Report	MediaWiki for browsing the report, page structure, and connected content.
Access the data	SPARQL · XML · APIs · Data dumps
Access full text	MediaWiki API · XSLT transformation