This project focuses on predicting the occurrence of fire or no-fire events in the Bejaia and Sidi Bel-Abbes regions of Algeria, based on weather data recorded from June 2012 to September 2012. The dataset includes weather attributes such as temperature, humidity, wind speed, and other fire-related indices.
# Fire Prediction in Algeria Using Weather Data
## Project Overview
This project focuses on predicting the occurrence of fire or no-fire events in the Bejaia and Sidi Bel-Abbes regions of Algeria, based on weather data recorded from June 2012 to September 2012. The dataset includes weather attributes such as temperature, humidity, wind speed, and other fire-related indices. The goal is to predict the likelihood of fire using various regression models.
The models used for training and prediction are:
- **Linear Regression**
- **Lasso Regression**
- **Ridge Regression**
- **Elastic Net**
Each of these models was tested to evaluate their performance on the dataset to predict whether a fire event occurred (fire class) or not (no-fire class).
## Dataset Information
The dataset consists of 244 instances, with 122 instances from each of the two regions (Bejaia and Sidi Bel-Abbes). These instances are classified into two categories:
- **Fire (138 instances)**
- **Not Fire (106 instances)**
### Dataset Period
- **Time Frame**: June 2012 to September 2012
- **Regions**: Bejaia (Northwest of Algeria) and Sidi Bel-Abbes
### Data Columns
The dataset includes 11 weather-related attributes and 1 output attribute (Class). The attributes provide critical information for predicting fire events. The attributes are:
1. **Date**: (DD/MM/YYYY) The date of weather observation, including the day, month (June to September), and year (2012).
2. **Temp**: The noon temperature (maximum temperature) in Celsius degrees, ranging from 22°C to 42°C.
3. **RH**: Relative Humidity in percentage, ranging from 21% to 90%.
4. **Ws**: Wind speed in km/h, ranging from 6 km/h to 29 km/h.
5. **Rain**: Total rainfall in mm, ranging from 0 mm to 16.8 mm.
6. **Fine Fuel Moisture Code (FFMC)**: Index from the Fire Weather Index (FWI) system, ranging from 28.6 to 92.5.
7. **Duff Moisture Code (DMC)**: Index from the FWI system, ranging from 1.1 to 65.9.
8. **Drought Code (DC)**: Index from the FWI system …