piuba-bigdata/contextualized_hate_speech

1---2language:3- es4pretty_name: contextualized_hate_speech5task_categories:6- text-classification7tags:8- hate_speech9size_categories:10- 10K<n<100K11---12# Contextualized Hate Speech: A dataset of comments in news outlets on Twitter13 14## Dataset Description15 16- **Repository: [https://github.com/finiteautomata/contextualized-hatespeech-classification](https://github.com/finiteautomata/contextualized-hatespeech-classification)** 17- **Paper**: ["Assessing the impact of contextual information in hate speech detection"](https://arxiv.org/abs/2210.00465), Juan Manuel Pérez, Franco Luque, Demian Zayat, Martín Kondratzky, Agustín Moro, Pablo Serrati, Joaquín Zajac, Paula Miguel, Natalia Debandi, Agustín Gravano, Viviana Cotik18- **Point of Contact**: jmperez (at) dc uba ar19 20### Dataset Summary21![Graphical representation of the dataset](Dataset%20graph.png)22This dataset is a collection of tweets that were posted in response to news articles from five specific Argentinean news outlets: Clarín, Infobae, La Nación, Perfil and Crónica, during the COVID-19 pandemic. The comments were analyzed for hate speech across eight different characteristics: against women, racist content, class hatred, against LGBTQ+ individuals, against physical appearance, against people with disabilities, against criminals, and for political reasons. All the data is in Spanish.23 24Each comment is labeled with the following variables25 26 27 28| Label      | Description                                                             |29| :--------- | :---------------------------------------------------------------------- |30| HATEFUL    | Contains hate speech (HS)?                                              |31| CALLS      | If it is hateful, is this message calling to (possibly violent) action? |32| WOMEN      | Is this against women?                                                  |33| LGBTI      | Is this against LGBTI people?                                           |34| RACISM     | Is this a racist message?                                               |35| CLASS      | Is this a classist message?                                             |36| POLITICS   | Is this HS due to political ideology?                                   |37| DISABLED   | Is this HS against disabled people?                                     |38| APPEARANCE | Is this HS against people due to their appearance? (e.g. fatshaming)    |39| CRIMINAL   | Is this HS against criminals or people in conflict with law?            |40 41 42 43There is an extra label `CALLS`, which represents whether a comment is a call to violent action or not.44 45The `HATEFUL` and `CALLS` labels are binarized by simple majority; the characteristic or category variables are put to `1` if at least one annotator marked it as such.46 47A raw, non-aggregated version of the dataset can be found at [piuba-bigdata/contextualized_hate_speech_raw](https://huggingface.co/datasets/piuba-bigdata/contextualized_hate_speech_raw)48### Citation Information49 50```bibtex51@article{perez2022contextual,52  author  = {Pérez, Juan Manuel and Luque, Franco M. and Zayat, Demian and Kondratzky, Martín and Moro, Agustín and Serrati, Pablo Santiago and Zajac, Joaquín and Miguel, Paula and Debandi, Natalia and Gravano, Agustín and Cotik, Viviana},53  journal = {IEEE Access},54  title   = {Assessing the Impact of Contextual Information in Hate Speech Detection},55  year    = {2023},56  volume  = {11},57  number  = {},58  pages   = {30575-30590},59  doi     = {10.1109/ACCESS.2023.3258973}60}61 62```63 64### Contributions65 66[More Information Needed]