# Wikipedia Cultural Diversity Dataset: A Complete Cartography for 300 Language Editions

arXiv: https://arxiv.org/abs/1901.07999 · HTML: https://mlsanity.com/paper/1901.07999/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Jan 2019.

## Built on by (most cited first, 4 of 4)

- [Probing Pre-Trained Language Models for Cross-Cultural Differences in Values](https://mlsanity.com/paper/2203.13722/) (2022)
- [SCoFT: Self-Contrastive Fine-Tuning for Equitable Image Generation](https://mlsanity.com/paper/2401.08053/) (2024)
- [LAG-MMLU: Benchmarking Frontier LLM Understanding in Latvian and Giriama](https://mlsanity.com/paper/2503.11911/) (2025)
- [Considerations for Multilingual Wikipedia Research](https://mlsanity.com/paper/2204.02483/) (2022)

## Abstract (arXiv)

In this paper we present the Wikipedia Cultural Diversity dataset. For each existing Wikipedia language edition, the dataset contains a classification of the articles that represent its associated cultural context, i.e. all concepts and entities related to the language and to the territories where it is spoken. We describe the methodology we employed to classify articles, and the rich set of features that we defined to feed the classifier, and that are released as part of the dataset. We present several purposes for which we envision the use of this dataset, including detecting, measuring and countering content gaps in the Wikipedia project, and encouraging cross-cultural research in the field of digital humanities.