Linear regression based prediction
# Algeria Forest Fires Dataset
## Overview
The dataset contains **244 instances** of forest fire occurrences from two regions in Algeria:
- **Bejaia Region** (Northeast Algeria)
- **Sidi Bel-abbes Region** (Northwest Algeria)
Each region contributes **122 instances**, covering the period **June 2012 to September 2012**. The dataset includes **11 attributes** along with **1 output class** that categorizes each instance as either **Fire (138 instances)** or **Not Fire (106 instances)**.
## Attribute Information
### **Date & Weather Observations**
- **Date**: (DD/MM/YYYY) - Ranges from June 2012 to September 2012
- **Temp**: Maximum temperature at noon (°C) - Range: `22 to 42`
- **RH**: Relative Humidity (%) - Range: `21 to 90`
- **Ws**: Wind speed (km/h) - Range: `6 to 29`
- **Rain**: Total rainfall in mm - Range: `0 to 16.8`
### **Fire Weather Index (FWI) System Components**
- **FFMC**: Fine Fuel Moisture Code - Range: `28.6 to 92.5`
- **DMC**: Duff Moisture Code - Range: `1.1 to 65.9`
- **DC**: Drought Code - Range: `7 to 220.4`
- **ISI**: Initial Spread Index - Range: `0 to 18.5`
- **BUI**: Buildup Index - Range: `1.1 to 68`
- **FWI**: Fire Weather Index - Range: `0 to 31.1`
### **Output Class**
- **Fire**: 138 instances
- **Not Fire**: 106 instances
## Usage
This dataset can be used for:
- **Predicting forest fire occurrences** based on weather conditions.
- **Developing machine learning models** for fire risk assessment.
- **Analyzing weather patterns** leading to fire incidents.