Machine learning project for predicting forest fire risk using meteorological data and Fire Weather Index (FWI) features.
# Algerian_Forest_Fire
# π₯ Algerian Forest Fires Prediction
---
## π Project Overview
This project focuses on predicting the occurrence of forest fires using the **Algerian Forest Fires Dataset**. The dataset contains meteorological and Fire Weather Index (FWI) features collected from two regions in Algeria.
The goal is to build a **robust classification model** that can distinguish between:
- π₯ Fire
- β
Not Fire
---
## π Dataset Summary
- **Total Samples:** 244
- **Regions:**
- Bejaia (Northeast Algeria)
- Sidi Bel-Abbes (Northwest Algeria)
- **Time Period:** June 2012 β September 2012
- **Target Classes:**
- Fire (138)
- Not Fire (106)
---
## π Dataset Features
### π¦οΈ Weather Data
- Temperature (Temp)
- Relative Humidity (RH)
- Wind Speed (Ws)
- Rain
### π₯ Fire Weather Index (FWI) Components
- FFMC
- DMC
- DC
- ISI
- BUI
- FWI
---
## π§ Problem Statement
Build a **machine learning classification model** that predicts whether a forest fire will occur based on environmental and weather conditions.
---
## βοΈ Tech Stack
- **Language:** Python
- **Libraries:**
- pandas
- numpy
- matplotlib
- seaborn
- scikit-learn
---
## π Project Workflow
1. Data Collection
2. Data Cleaning & Preprocessing
3. Exploratory Data Analysis (EDA)
4. Feature Engineering
5. Model Training
6. Model Evaluation
7. Prediction
---
## π Model Performance
| Model | MAE | RΒ² Score |
|---------------------|--------|----------|
| Linear Regression | 0.5468 | 0.9848 |
| Lasso Regression | 0.6200 | 0.9821 |
| Ridge Regression | 0.5642 | 0.9843 |
| ElasticNet | 0.6576 | 0.9814 |
> *Metrics used: Mean Absolute Error (MAE) and RΒ² Score*
> *Note: Performance may vary based on preprocessing and tuning.*
---
## π§Ή Data Preprocessing
- Handled missing values
- Encoded categorical variables
- Feature scaling (Standardization)
- Date feature extraction
---
## π Future Improvements
- Hyperparameter tuning (GridSearchCV)
- Deployment using Flask β¦