metadata
license: mit
task_categories:
- text-classification
- token-classification
language:
- en
Hate and Offensive Speech Dataset
This dataset was created using several datasets that can be found on Hugging Face:
-SetFit/hate_speech_offensive:https://huggingface.co/datasets/SetFit/hate_speech_offensive -tweets_hate_speech_detection:https://huggingface.co/datasets/tweets_hate_speech_detection -thefrankhsu/hate_speech_twitter:https://huggingface.co/datasets/thefrankhsu/hate_speech_twitter -parnoux/hate_speech_open_data_original_class_test_set:https://huggingface.co/datasets/parnoux/hate_speech_open_data_original_class_test_set
Dataset Details
The dataset contains columns:
-tweets_cleaned - that stores tweets, cleaned to some extent, may need additional work for some tasks
-label - data labels, where 0 means hate, 1 means offensive speech, 2 means neither