ML Wine Quality Predictor

A completed machine learning project that predicts wine quality ratings (0-10) from physicochemical properties such as acidity and alcohol content. It applies multinomial logistic regression with exploratory data analysis across red and white wine datasets and identifies the features most predictive of quality through comparative model evaluation.

Highlights

• Multinomial logistic regression for multi-class wine quality classification
• Exploratory data analysis and feature engineering on physicochemical datasets
• Model evaluation with cross-validation, accuracy, precision, and recall
• Feature-importance analysis to surface the strongest quality predictors
• Standardized preprocessing pipeline for both red and white wine varieties