A simple NLP model used to detect fake news written in Tagalog.
To run the inference and test the model, open the Inference.ipynb notebook. It contains instructions on how to load the model, preprocess the input, and make predictions.
The model uses a binary classification threshold:
- Score ≥ 0.5 → Fake news
- Score < 0.5 → Non-fake news
This project uses an open-source dataset from jcblaisecruz02.