Skip to content

Project 5: Ensemble Machine Learning - Wine Quality Prediction

Overview

This project implements and compares ensemble machine learning models to predict wine quality categories using the UCI Wine Quality Dataset.

Dataset

Models Evaluated

  • Random Forest (100 trees): Bagging approach with parallel tree training
  • Gradient Boosting (100 trees): Sequential boosting with error correction

Key Results

Model Test Accuracy Test F1 Score Accuracy Gap
Random Forest 88.75% 0.8661 0.1125
Gradient Boosting 85.62% 0.8411 0.1039

Files

  • ml05_webb.ipynb: Main Jupyter notebook with analysis
  • winequality-red.csv: Wine quality dataset
  • README.md: This file

Technologies Used

  • Python 3.x
  • pandas, numpy, matplotlib
  • scikit-learn (ensemble methods, metrics)

Author

Adrianna Webb - November 23, 2025