---
license: mit
task_categories:
- text-classification
language:
- id
size_categories:
- 10K<n<100K
---

This dataset is taken from nlp-brin-id/id-hoax-report-merge-v2. </br>
The subsets can be utilized as samples for unsupervised contrastive learning based on longer textual claim (content of the public report based on the corresponding title of the claim) as input for fake news detection. </br> 

Attributes used = 'Content'.</br>