Ridge vs Lasso Regression Comparison

📌 Project Overview

Performance comparison of Ridge (L2) and Lasso (L1) regression models on the Diabetes dataset.
Objective: Identify the most effective regularization method for blood glucose level prediction.

🚀 Key Features

• Data preprocessing pipeline
• Hyperparameter optimization with GridSearchCV
• Coefficient analysis and visualization
• Model performance comparison

📊 Dataset Information

Source: diabetes_clean.csv (768 samples)
Target Variable: glucose (blood glucose level)

Key Features:

• pregnancies
• diastolic (blood pressure)
• triceps (skin thickness)
• diabetes (diagnosis)
• age
• bmi

🧠 Model Details

Optimized Parameters

Model	Best Alpha
	1
	0.1

Coefficient Comparison

Feature	Ridge	Lasso
diabetes	24.87	24.60
dpf	1.70	0.81
age	0.49	0.49
pregnancies	-0.46	-0.45

📈 Key Results

• Ridge: Balanced shrinkage while retaining all features
• Lasso: 52% coefficient reduction for dpf feature
• Common Finding: diabetes is the strongest predictor

🛠️ Installation

git clone https:/barisgudul/Ridge_vs_Lasso_Analysis.git
cd Ridge_vs_Lasso_Analysis

🖥️ Usage

# Run the analysis script
jupyter notebook LvsR.ipynb

📄 License

MIT License

Permissions:
✅ Free academic/research use
✅ Modification and redistribution
❌ Commercial use requires written consent

Full license terms available in LICENSE file.

---

📧 Contact Information

Project Maintainer

Contribution Guidelines:
We welcome collaborations! Please reach out via email before submitting PRs.