Harmonizing existing climate change mitigation policy datasets with a hybrid machine learning approach.

Area of Science:

• Environmental Science
• Climate Policy Analysis
• Data Science

Background:

• Increasing number and scope of global climate policies necessitate comprehensive data.
• Existing datasets lack multi-indicator information on policy elements and implementation contexts.
• Demand for a global-level dataset to track climate change mitigation efforts.

Purpose of the Study:

• To develop the Global Climate Change Mitigation Policy Dataset (GCCMPD).
• To provide a multi-indicator dataset on climate policies and their implementation.
• To facilitate a deeper understanding of climate activities globally.

Main Methods:

• Utilized a semisupervised hybrid machine learning approach.
• Integrated policy information from global, regional, and sector-specific sources.
• Employed expert knowledge-based dictionary mapping, probability statistics, and natural language processing.

Main Results:

• Developed the GCCMPD, containing 73,625 policies from 216 entities.
• Classified multiple indicators including objectives, target sectors, instruments, and legal compulsion.
• Aligned dataset with the Intergovernmental Panel on Climate Change (IPCC) emission sector classification.

Conclusions:

• The GCCMPD offers detailed and consistent information on climate mitigation policies.
• The dataset aids policymakers, researchers, and organizations in analyzing climate actions.
• Facilitates comparison of climate activities across countries, sectors, and entities.