datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
bangla-fake-news
Bangla Fake News Detection Dataset
One of the first publicly available fake news datasets for the Bengali language,
scraped from local newspapers. Built to support NLP research in under-represented languages.
Dataset Structure
Collected from Bengali local news sources
Labeled as fake / real
Includes a Bengali stemmer and corpus builder
Benchmark Results
Model
Accuracy
Naïve Bayes
52%
Logistic Regression
77%
Random Forest
85%… See the full description on the dataset page: https://huggingface.co/datasets/Far121/bangla-fake-news.bangla-news-articles-sample