Revenue Prediction using Machine Learning

Overview

This project predicts restaurant revenue using operational, categorical, and location features.

Techniques Used

• Data cleaning & preprocessing
• One-hot encoding & feature scaling
• Log transformation for skewed target
• Multiple regression models
• Random Forest with cross-validation
• Feature importance & residual analysis

Results

• Best Model: Random Forest Regressor
• Cross-validated R² ≈ 0.89

Business Insights

• Order frequency is the primary driver of revenue
• Franchise presence, city, and category significantly impact revenue

4. Follow the notebooks in order from exploratory analysis to final model.

---

📌 Insights

✔ Best practical model: Random Forest
✔ Most influential feature: Order count
✔ Secondary drivers: franchise, city, category

---

🛠 Tools & Libraries

• Python
• Pandas, NumPy
• Scikit-learn
• Matplotlib, Seaborn