Source Code A Hybrid Machine Learning Framework for Interpretable Wine Quality Assesment with SHAP-based Explainability

The study utilizes the UCI Wine Quality dataset (red and white Vinho Verde wines) to develop an interpretable hybrid machine learning framework for binary wine quality prediction. Data preprocessing includes feature engineering and feature selection prior to model development. Several baseline classifiers and ensemble approaches are evaluated, with the Soft Voting Classifier demonstrating the best predictive performance. The repository includes: 1. Jupyter notebooks for data preprocessing, feature engineering, model training, evaluation, and SHAP analysis2. Streamlit dashboard source code for interactive prediction and explainability3. Trained machine learning models and preprocessing artifacts4. Original and processed datasets5. Evaluation results and visualization outputs6. Python dependency list for reproducing the experiments The repository is intended to support transparency, reproducibility, and future research in interpretable machine learning for wine quality prediction