keshavpal06/Algerian-Forest-Fire-Regression

# Algerian Forest Fire Regression Project This repository hosts an end-to-end Machine Learning pipeline that predicts the **Fire Weather Index (FWI)** using meteorological conditions and underlying index markers. The project treats this ecological problem as a regression task, applying data cleaning, Exploratory Data Analysis (EDA), feature engineering, model training, and regularization techniques to achieve highly accurate predictions. Additionally, it features a deployed **Flask Web Application** to provide real-time interactive model inference. --- ## 📌 Project Overview Forest fires present a major risk to human lives, properties, and biodiversity. Predicting the Fire Weather Index (FWI)-a key indicator used to estimate fire intensity and behavior-can empower forestry departments to deploy preemptive safety measures. This project automates that assessment by training regression algorithms on real-world climatic measurements. --- ## 📊 Dataset Information The dataset used in this project originates from the **UCI Machine Learning Repository** and captures observations across two specific regions in Algeria: 1. **Bejaia Region** (Northeast Algeria) - 122 instances 2. **Sidi Bel-Abbes Region** (Northwest Algeria) - 122 instances **Timeline Covered:** June 2012 to September 2012 **Total Samples:** 244 instances ### Attribute Description #### Meteorological Variables: * **Date:** Day, month (`June` to `September`), and Year (`2012`). * **Temperature (Temp):** Noon max temperature in Celsius degrees (Range: 22 to 42°C). * **Relative Humidity (RH):** Humidity percentage value (Range: 21% to 90%). * **Wind Speed (Ws):** Wind velocity measurement in km/h (Range: 6 to 29 km/h). * **Rain:** Total precipitations of the day in mm (Range: 0 to 16.8 mm). #### Fire Weather Index (FWI) Components: * **FFMC (Fine Fuel Moisture Code):** Numeric index evaluating moisture content of litter and ignition potential (28.6 to 92.5). * **DMC (Duff Moisture Code):** Numeric index d …