HOT Speech: Hateful, Offensive, Toxic Speech Dataset
Twitter (X)
YouTubeA HOT Speech (hate/offensive/toxic speech dataset) dataset of 3,481 social media comments with multi-dimensional annotations, collected by SOMAR / ICPSR via Twitter, YouTube, and Reddit APIs (August 2021). Each record contains comment text, hard and soft labels (hateful, offensive, toxic), composite scores, and 21 researcher codes. multi-platform dataset with rich annotations including both MTurk worker labels and researcher codes. Suitable for hate speech detection, offensive language identification, toxicity classification, and political discourse analysis.
Cite this dataset at
hot-speech.echo.directory
Analytics
Topic 0
Topic 1
Topic 2
Topic 3
Topic 4
Topic 5
Topic 6
Topic 7
Topic 8
Topic 9
Topic 10
Topic 11
Show all 19 topics
Topic 12
Topic 13
Topic 14
Topic 15
Topic 16
Topic 17
Topic 18
Computed over the full corpus.
Record
- Catalog key
hot-speech- Authors
- Resnick, Paul J.; Hemphill, Libby; Wu, Siqi; Schöpke-Gonzalez, Angela; Kumar, Sagar
- Year
- 2023
- Source paper
- HOT Speech: Comments from Political News Posts and Videos that were Annotated for Hateful, Offensive, and Toxic Content
- DOI / arXiv
- 10.3886/45fc-9c8f
- License
- Not stated (ICPSR SOMAR terms of use apply)
- Records
- 3,481
- Text corpus
- comment text
- Languages
- en
- Collected
- 2021-08 (2020s)
- Tasks
- Toxicity detection (multi-label;
offensive_hard, toxic_hard, offensive_soft, toxic_soft, composite_offensive, composite_toxic, code_insult, code_threat) inferred, Hate speech detection (multi-label;hateful_hard, hateful_soft, composite_hateful) inferred, Fake news detection (binary;code_misinformation) inferred - Region
- Global (inferred: the platform is used worldwide)
Provenance
- Originally published at
- https://doi.org/10.3886/45fc-9c8f link checked
- Hosted copy
- Mirrored on echo.directory on 2026-10-09, 1.5 MB across 2 files. Checksums in MANIFEST.json.
How to cite
BibTeX
@misc{echo2026,
title = {ECHO: Exploratory Compendium of Humanity Online},
author = {Switzer, Braden and Zafarani, Reza},
year = {2026},
url = {https://echo.directory},
note = {Syracuse University}
}
@misc{https://doi.org/10.3886/45fc-9c8f,
doi = {10.3886/45FC-9C8F},
url = {https://www.icpsr.umich.edu/sites/somar/view/studies/300473/versions/V1},
author = {Resnick, Paul J. and Hemphill, Libby and Wu, Siqi and Schöpke-Gonzalez, Angela and Kumar, Sagar},
keywords = {news media, politics, YouTube, Reddit},
language = {en},
title = {HOT Speech: Comments from Political News Posts and Videos that were Annotated for Hateful, Offensive, and Toxic Content},
publisher = {ICPSR - Interuniversity Consortium for Political and Social Research},
year = {2026}
} APA
Switzer, B., & Zafarani, R. (2026). ECHO: Exploratory Compendium of Humanity Online [Data repository]. Syracuse University. https://echo.directory
Resnick, P. J., Hemphill, L., Wu, S., Schöpke-Gonzalez, A., & Kumar, S. (2023). HOT Speech: Hateful, Offensive, Toxic Speech Dataset [Data set]. https://doi.org/10.3886/45fc-9c8f
MLA
Switzer, Braden, and Reza Zafarani. ECHO: Exploratory Compendium of Humanity Online. Syracuse University, 2026, echo.directory.
Resnick, Paul J., et al. HOT Speech: Hateful, Offensive, Toxic Speech Dataset. 2023, doi.org/10.3886/45fc-9c8f.