POLIANNA: Annotated EU Climate Policy Texts for Machine Learning

20,577 annotated spans across 412 articles from 18 EU climate change mitigation and renewable energy laws. The dataset was created by researchers including Sebastian Sewerin from ETH Zurich and published in 2023. It provides a novel coding scheme for policy design concepts to assist in scaling policy analysis through supervised machine learning.

Use Cases

• Train supervised models for automatic paragraph identification based on annotated policy design concepts.
• Scale manual coding of policy texts using machine learning approaches as described.
• Calculate inter-annotator agreement measures for text annotation projects.
• Process text annotations generated by the INCEpTION tool using the provided code.

Strengths

• Contains 20,577 annotated text spans, providing substantial labeled data.
• Draws from 412 articles across 18 specific EU laws, offering a focused corpus.
• Includes code for calculating inter-annotator agreement and processing annotations.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

ETH Zurich

Collection Method

Annotated legislative texts from the European Union.

Geography

European Union