ManelHjaoujia/Algerian-Forest-Fires-Regression
Domain:
environment and energyCreator:
Man
Host:
Regression analysis on Algerian Forest Fires dataset using Linear, Lasso, Ridge, and ElasticNet.
# Algerian Forest Fires Regression
This project applies different regression techniques to predict the Fire Weather Index (FWI) using the **Algerian Forest Fires dataset**.
## Steps
1. Data cleaning & preprocessing
2. Feature encoding & scaling
3. Correlation analysis & multicollinearity check
4. Model training:
- Linear Regression
- Lasso Regression (+ Cross Validation)
- Ridge Regression (+ Cross Validation)
- ElasticNet Regression (+ Cross Validation)
5. Model evaluation with MAE and R²
## Models
- **Linear Regression**: baseline
- **Lasso**: performs feature selection
- **Ridge**: reduces multicollinearity
- **ElasticNet**: combines Lasso & Ridge
## Dependencies
- pandas, numpy
- matplotlib, seaborn
- scikit-learn