A machine learning project on the Algerian Forest Fires dataset involving data cleaning, visualization, and regression modeling using Lasso, Ridge, and ElasticNet.
# Project2_algerian_forest_prediction
A machine learning project on the Algerian Forest Fires dataset involving data cleaning, visualization, and regression modeling using Lasso, Ridge, and ElasticNet.
# Algerian Forest Fires Regression Analysis 🌲🔥
This project analyzes the **Algerian Forest Fires dataset** using machine learning techniques to model and predict fire occurrences based on meteorological and regional features. The analysis includes **data cleaning**, **visualization**, and **regression modeling** using **Lasso**, **Ridge**, and **ElasticNet**—along with their **cross-validated versions**.
---
## 🔍 Project Objectives
- Clean and preprocess the Algerian Forest Fires dataset
- Visualize key patterns and relationships in the data
- Apply and compare multiple regression models:
- Lasso Regression
- Ridge Regression
- ElasticNet Regression
- LassoCV, RidgeCV, ElasticNetCV (cross-validation)
- Evaluate models using performance metrics like MAE and R² Score
---
## 🛠️ Technologies Used
- Python 3.x
- NumPy, Pandas
- Matplotlib, Seaborn
- Scikit-learn
---
## 📊 Visualizations
Exploratory Data Analysis (EDA) includes:
- Correlation heatmaps
- Pairplots
- Feature distributions
- Fire occurrence patterns
---
## 🤖 Machine Learning Models
| Model | Regularization | Cross-Validated Version |
|-------------------|----------------|--------------------------|
| Lasso | ✅ | ✅ (`LassoCV`) |
| Ridge | ✅ | ✅ (`RidgeCV`) |
| ElasticNet | ✅ | ✅ (`ElasticNetCV`) |
Each model is evaluated using:
- **R² Score**
- **Mean Absolute Error (MAE)**
---