README.md

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


<b>We do not maintain this repository further. For accessing the most recent Indonesian Fake News dataset that we created, please visit BRIN's dataverse: </b> <url>https://data.brin.go.id/dataset.xhtml?persistentId=hdl:20.500.12690/RIN/7QBRKQ</url></br></br>


We create this dataset from nlp-brin-id/id-hoax-report-merge-v2. </br>

In this subset, triplets candidates are described as 
Positive sentence A; Positive Sentence B; Negative sentence C. Sampling space are defined as follows:</br>

- [HOAX label] Title Hoax; Title Hoax; Title Non-Hoax 
- [HOAX label] Title Hoax; Title Hoax; Content Non-Hoax (if not empty)  
- [HOAX label] Title Hoax; Content Hoax (if not empty); Title Non-Hoax - 
- [HOAX label] Title Hoax; Content Hoax (if not empty); Content Non-Hoax (if not empty) 
- [HOAX label] Title Hoax; Title Hoax; Fact Hoax (if Fact != null), 
- [HOAX label] Title Hoax; Content Hoax (if not empty); Fact Hoax (if Fact != null), 
- [NON-HOAX label] Title Non-Hoax; Title Non-Hoax; Title Hoax 
- [NON-HOAX label] Title Non-Hoax; Title Non-Hoax; Content Hoax 
- [NON-HOAX label] Title Non-Hoax; Content Non-Hoax (if not empty); Title Hoax 
- [NON-HOAX label] Title Non-Hoax; Content Non-Hoax (if not empty); Content Hoax (if not empty) 
- [NON-HOAX label] Title Non-Hoax; Fact Non-Hoax (if not empty); Title Hoax 
- [NON-HOAX label] Title Non-Hoax; Fact Non-Hoax (if not empty); Content Hoax  (if not empty) 
- [NON-HOAX label] Content Non-Hoax (if not empty); Fact Non-Hoax (if not empty); Title Hoax 
- [NON-HOAX label] Content Non-Hoax (if not empty); Fact Non-Hoax (if not empty); Content Hoax (if not empty)

For creating the subset, we permute hard negative samples for 10 epochs dependent to the class category. </br>
For each epoch, we flip coins to decide whether the triplet uses 'Title', 'Content' (a long description of claim in Title), or 'Fact'.</br>
Note that: 'Fact' represents hard negative or contradicting sentence in Hoax class samples, while in Non-Hoax subset it represents supports (Positive sentence). </br>