This project applies four machine learning models (Logistic Regression, Decision Trees, Random Forests, and KNN) to classify individuals into obesity categories based on demographic and lifestyle features.
Source: [UCI Machine Learning Repository – Obesity Dataset]
Variables include age, gender, weight, height, physical activity, eating habits, and other health-related factors.
Multinomial Logistic Regression
Decision Tree
Random Forest
K-Nearest Neighbors (KNN)
Random Forest: Best accuracy (93.8%)
KNN (k=3): Strong performance (~83%)
Logistic Regression: Moderate performance but interpretable
Decision Tree: Lower accuracy but useful for variable importance
Random Forest is recommended for predictive performance, while Logistic Regression is suitable for explainability