ISAAC is a corpus of 527,060,919 English-language Reddit posts from 2007 to 2023 about six social group distinctions: ability, age, race, sexuality, skin tone and body weight. Every post carries 59 fields, including labels for moralization, sentiment, emotion, linguistic generalization and estimated user location.
| Corpus (341 GB, one configuration per social group) | BabakScrapes/isaac-reddit |
| Classifiers (9 models) | the repositories on this page |
| Try the classifiers without code | BabakScrapes/isaac-classifiers |
| Project website, browsing and direct download | isaac.psychology.illinois.edu |
| Pipeline source, keyword lists, pattern sets | GitHub |
| Python loader | isaac-data |
| Data Use Agreement | Data_Use_Agreement.md |
The corpus is hosted on a personal account rather than this organization for storage reasons only. It is the same project and the same terms.
The nine classifiers on this page are released under a Creative Commons Attribution 4.0 International License, without access restrictions. The corpus is governed by the ISAAC Data Use Agreement, which prohibits re-identification and redistribution. The location model is not published here: access can be requested by email under the ISAAC Model Use Agreement.
One citation covers the whole project, including the corpus, the pipeline, the models and every access route. Please do not cite the repositories separately.
@article{hemmatian2026isaac,
author = {Hemmatian, Babak and Hadjarab, Sarah and Chen, Jessica and Kurdi, Benedek},
title = {The {Illinois} Social Attitudes Aggregate Corpus ({ISAAC}): An Open Tool and Reproducible Pipeline for Analyzing Social Group Discourse at Scale},
year = {2026},
journal = {arXiv},
eprint = {2609.27059},
archivePrefix = {arXiv},
primaryClass = {cs.CL},
doi = {10.48550/arXiv.2609.27059},
url = {https://arxiv.org/abs/2609.27059}
}