About the project
An end-to-end machine learning study on the Wine Quality dataset. Nine classifier variants — Decision Tree, k-NN and Random Forest, three hyperparameter settings each — are evaluated with stratified 5-fold cross-validation, alongside a VotingClassifier ensemble that aggregates all nine by majority vote.
What it demonstrates
- Multiclass target
- Nine classifier variants
- Majority-voting ensemble
- Leakage-free by construction
- Two metrics, deliberately
Stack
- Python
- scikit-learn
- pandas
- Plotly
- Streamlit
