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

Raw dataset for "Fact-Aware Fake-news Classification for Indonesian Language"</br></br>
<b>Disclaimer:</b> Beta version, contains imbalanced representation of domain-specific NON-HOAX samples. We will release a new training and evaluation suite soon as a replacement of this dataset. </br></br>

Data originates from https://turnbackhoax.id/ (Mafindo data 2018-2023); </br>
The attributes of data are: </br>
1. Label_id: Binary class labels ("HOAX"==1 ; "NON-HOAX"==0).</br>
2. Label: Binary class labels ("HOAX" or "NON-HOAX").</br>
3. Title: Claim or headline of news article.</br>
4. Title_cleaned: Preprocessed claim, by removing tag label at the beginning of the sentence.</br>
5. Content: the content of news article. </br>
6. Fact: The summary of factual evidence that is either supporting or contradicting the correponding claim.</br>
7. References: URL link of news article and the corresponding verdict or factual evidence as the justification of the news article.</br>
8. Classification: Fine-grained classification labels for the news article:</br> 'CekFakta', 'Fabricated Content', 'False Connection',
       'False Context', 'Impostor Content', </br> 'Manipulated Content',
       'Misleading Content', 'Satire', 'nan'.</br></br>

Example of usage:</br>
```python
>>> from datasets import load_dataset
>>> train_dataset = load_dataset(
...     "nlp-brin-id/fakenews-id-brin",
...     split="train",
...     keep_default_na=False,
... ).select_columns(['Label_id', 'Title_cleaned', 'Content', 'Fact'])
```