unimelb-nlp/

MultiEup-v2

This dataset card documents Multi-EuP-v2, a multilingual corpus of European Parliament debate speeches enriched with Member of European Parliament (MEP) metadata and multilingual debate titles/IDs. It supports research on political text analysis, speaker-attribute prediction, stance/vote prediction, multilingual NLP, and retrieval.

	Dataset Details




	
	Dataset Description

Multi-EuP-v2 aggregates 50,337 debate speeches (each a unique did) in 24… See the full description on the dataset page: https://huggingface.co/datasets/unimelb-nlp/MultiEup-v2.