Skip to content

Dataset · Davlan

Multilingual SIB-200 topic classification dataset

Davlan/sib200

Largest open corpus for topic classification of Wikipedia sentences across 205 languages and dialects.

What's inside

A text dataset of 100k to 1M examples built on Flores-200 and covering 205 languages and dialects. It contains Wikipedia sentences labeled by topic (e.g., science and technology, sports, politics), with ready train, validation, and test splits for each language.

What you can do with it

  • Training and evaluating multilingual text classifiers, including rare and low-resource languages
  • Benchmarking cross-lingual transfer quality of language models and embeddings
Tasks
Text classification
Languages
Acehnese, acm, acq, Tunisian Arabic, Afrikaans, apc, Akan, Albanian and 4 more
Size
100K–1M rows
Downloads this month
19.2K
Likes
25

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("Davlan/sib200")