Culture-and-Morality-Lab/CAMEL_Dataset

Description

The Cultural and Moral Expressions in Language (CAMEL) corpus is a dataset of more than 56000 texts gathered from multiple sources including Reddit, X (Twitter), Wikipedia, and other relevant text datasets. This dataset was prepared by the Culture and Morality Lab (CaM-L) at the University of Massachusetts Amherst. For an overview of our work, visit us here 🐪

Each piece of text has been annotated by at least three human annotators. The annotators received intensive training in recognizing 25 constructs related to cultural and social psychology. Examples of the key constructs include moral foundations (Haidt & Joseph, 2004; Atari et al., 2023), kinship (Schulz et al., 2019), and cultural tightness-looseness (Gelfand, Nishii, & Raver, 2006). Please note that the dataset contains offensive or hateful language as well, which has been annotated in the appropriate category.

License

This dataset is governed by the Creative Commons Attribution Non Commercial License 4.0 (CC-BY-4.0). You may freely use the dataset for non-commercial purposes, as long as you properly attribute/cite the relevant source (see below).

Citation Information

@misc{zewail2025camel,
  author       = {Aliah Zewail and Aanchal Setia and Roya Mohammadsadegh and Fırat Şeker and Ali Hajian and Hector J. Sosa and Satvika Reddy Gavireddy and Liora Morhayim and Mohammad Atari},
  title        = {The Cultural and Moral Expressions in Language (CAMEL) Corpus},
  year         = {2025},
  note         = {Unpublished manuscript}
}

Contact

For questions or comments about this dataset, please reach out to Mohammad Atari (matari@umass.edu) or Aliah Zewail (azewail@umass.edu).