A project from #semanticClimate: https://semanticclimate.github.io/p/en/.
The Climate Knowledge Graph (ClimateKG) is in Alpha. We are looking for support to take the project into full production. Get in contact or check our discussion board.
We envisage equitable access to climate change science by indexing climate science literature into the ClimateKG.
Problem: Climate science literature and policy reports are not easily searchable and reusable. Climate Change is now a crisis and needs to be treated as an emergency — some of the few solutions and hope for the survival of humanity are to be found in science — but current knowledge management systems are not fit for purpose.
Solution: Create the Climate Knowledge Graph (ClimateKG) mapping the UN Climate Change literature, Open Access climate literature, and Grey Literature — policy and technical reports, etc.
- Make the ClimateKG: Currently it includes 70 chapters of UN IPCC AR6, semantic identifiers for 100,000+ paragraphs, 2000 climate terms aligned with Wikidata, and a 600+ term glossary.
- Open-source: App for search, collation, and reuse; Software for adding corpora to ClimateKG.
- Community and Citizen Science: Feb. 2024, >300 participants in 13 in-person events.
The web has a design fault — it has no cataloguing or indexing system. The web’s creator Tim Berners-Lee, later, proposed a solution ‘the Semantic Web’. But it was too late, it didn’t take off, and since then the rest of us have had to work with second best. The tech giants exploited the fault and created their wealth by making their own private indexes. An example of the Semantic Web is Google Knowledge Graph which has 500 billion facts (2020) and generates the infoboxes on search results (https://en.wikipedia.org/wiki/Google_Knowledge_Graph).
Knowledge Graphs are part of the modern implementation of the Semantic Web — which enables automated knowledge management using data about definitions and relationships.
The Climate Knowledge Graph is: a modern catalogue and index of scientific literature and Grey Literature — policy reports, technical documents, etc. It is open-source and Open Science based. The technology allows for search, semantification (adding meaning), data analysis, machine learning, and AI (scientifically validated, Open Science compliant, and ethically based in SDGs). The knowledge graph creates a map of the literature as a data set — that can be used for granular search and can retrieve results down to a sentence level from documents anywhere on the web. It also assigns meaning of content by using a combination of ‘definitions’ and ‘relationships’. Definitions use controlled vocabularies like for ‘Organization’ (https://schema.org/Organization). A relationship is that an Organisation is ‘Author’ (https://schema.org/author) — such as the UN IPCC.
With only these two bits of data: Author is Organisation UN IPCC — data analysis can be applied to retrieve information about the author, for example: What type of organisation is it? An NGO, or; What was the annual budget of the author for 2020? 8,627,688 CFH.