IndoNewsClassifier
A text classifier for Indonesian-language news articles, deployed as a live Streamlit demo.
− reported only the flattering accuracy
+ accuracy + macro-F1 for both models
+ TF-IDF 82.8/.80 · IndoBERT 89.7/.87
+ fixed an empty-input crash
Overview
A machine learning model that classifies Indonesian-language news articles by category, wrapped in a Streamlit app so it's actually usable rather than just a notebook.
Technologies Used
- Python, scikit-learn — text classification pipeline
- Streamlit — deployed, interactive demo
- NLP preprocessing for Bahasa Indonesia text
Why it's featured
Out of the projects in my GitHub, this is the one that needed the least cleanup — the training methodology, the reported results, and the deployed demo all line up with each other. It's the baseline I held the other projects to.