Skip to content

Data science and AI for researching climate change policies and laws

Using leading data science methods and artificial intelligence, we’re building the world’s largest open knowledge base for climate law and policy. Our goal is to provide useful and accessible big data and insights to inform climate action.

Open data and tools for climate policy

Our open, cost-free and AI-powered tool is the first of its kind. It allows you to quickly and easily search through the complete text of thousands of climate change interventions, laws, policies and litigation cases from every country. Use our data to identify climate policy gaps, opportunities for action and examples of best practice.

Our data

Our current climate policy database consists of thousands of laws, policies, strategies and action plans from all national governments.

The data are sourced in collaboration with the Grantham Research Institute on Climate Change and the Environment at LSE.

We are working to add additional documents, including submissions to the UNFCCC, and laws and policies from sub-national and local governments. We will source these by:

1

Collaborating with international, sectoral and regional database providers.

2

Extracting law, policy and strategy documents (‘scraping’) from government and other websites.

3

Crowd-sourcing documents.

If there is a dataset you think we should include, please get in touch.

Data science and AI

We use innovative data science and AI methods to turn masses of unstructured text from climate policy and legal documents into structured, usable and useful information.

text-icon

Extracting text: document parsing

We are building and training machine learning models that can automatically read and extract text from PDFs and websites, enabling us to structure and share information from thousands of law and policy documents.

This will mean we can massively ramp up the speed of data collection as we find new data sources. Access our GitHub page for more information.

Semantic Search Icon

Natural language search

Our application of machine learning is tackling blind spots in policy research by making it quicker, easier and more intuitive to search climate policies and laws.

Our tool identifies similar and related terms to search queries, so there is no need to rely on typing specific keywords into the search bar to find relevant information. Find out how our natural language search works.

Text Magnifier Icon

Text classification and information retrieval

We are training machine learning models to identify and link useful concepts across the texts of thousands of documents. This will allow understanding of, for example, how governments address different extreme weather events, or regulate various environmental technologies.

Using existing taxonomies and new bespoke ones, we will help identify concepts quickly, pinpoint trends and highlight gaps.

Doc Switch Icon

Multi-language approach

Search in English, find results in all languages: We are working towards allowing our users to input a search term in English but get results from documents that are published in different languages, plus an English translation.

User interface in other languages: We will use machine translation to offer our tool in other languages, including all six official UN languages.

Knowledge Icon

Open knowledge graph

We are working towards an open knowledge graph that joins the dots between points in our growing structured dataset, revealing gaps, risks and opportunities.

This will open up the ability to explore our data in new ways, and connect it to other datasets and APIs to start to measure the impacts of policy decisions.

Graph Icon

Analytical tools

We will develop tools to allow stakeholders to generate their own insights, including linking our data with ‘real-world’ external data like carbon emissions.

This will make it easier to link actions to outcomes, and facilitate modelling of policy effectiveness and risk analyses.

Sharing knowledge and resources

All our data are free for anyone to access and use through our web tool. We are also working towards releasing a set of open APIs so that individuals and organisations can use our codes in their own programmes of work.

Soon we will launch data visualisations to make it easier to explore and contextualise trends in data, working towards analyses and publication of our insights.

We regularly post updates about our progress on our blog and changelog.

Our users

The data and knowledge we generate are open and free to everyone. But we believe the people who will benefit from them the most are:

Policymakers

Use our tool to explore ‘menus’ of policy options and find out what climate change interventions are being implemented where and by whom, and identify best practice to shape policy design.

Researchers

Explore our data to identify different climate change policies, policy instruments and climate targets worldwide; model the potential impact of climate laws and policies and carry out cost-benefit analyses; and evaluate the effectiveness of climate change strategies.

Citizens and civil society

Access our information to guide voting choices, question decisions, highlight insufficient action and suggest alternative ways forward. Our data will help strengthen strategic legal cases to challenge governments and corporations and claim damages for the consequences of climate change.

Private sector

Apply our data to integrate relevant laws and policies into risk modelling. Our data will also lead to a better understanding of the regulatory environment, guiding smarter investment decisions, climate finance deployment and responsible lending.

We are open data, open source

  • Justice Scales Icon Democratises access to information and promotes climate justice, as often those who need access to data the most are least able to pay for it.
  • Spring Icon Provides a springboard for innovation, accelerating research efforts and catalysing collaboration.
  • Middle Icon Corrects biases - by opening up our models to scrutiny and ensuring that we remain neutral and impartial.