🔥 Algerian Forest Fires Regression Analysis

Comparing Regression Models for Fire Likelihood Prediction

📖 Overview

This project compares various regression techniques to predict the likelihood of forest fires in Algeria using meteorological and environmental data from the UCI Machine Learning Repository.

🛠️ Tech Stack

📊 Dataset Description

• Dataset Name: Algerian Forest Fires Dataset
• Source: UCI Machine Learning Repository
• Features:
  • 🌡️ Meteorological: Temperature, Humidity, Wind Speed, Rain
  • 🔥 FWI Components: FFMC, DMC, DC, ISI, BUI
  • 📅 Temporal: Day, Month
• Target: Fire likelihood (continuous variable)

🧠 Algorithms Tested

• 📈 Linear Regression
• 📉 Polynomial Regression (Best: degree=2)
• 📍 K-NN Regression (Best: k=3, Manhattan distance)
• 🌳 Decision Tree (Best: unlimited depth)
• 🔄 SVR (Best: RBF kernel)

📈 Results

Model	Mean MSE	Min MSE	Max MSE
Linear Regression	0.0849	0.0572	0.1191
Polynomial Regression	0.4995	0.1268	1.0032
K-NN Regression	0.0380	0.0088	0.0546
Decision Tree	0.0143	0.0028	0.0256
SVR	0.0557	0.0284	0.0954

🚀 Quick Start

git clone https://github.com/JaskiratCodeKaur/RegressionAnalysis.git
cd algerian-forest-fires-regression
python main.py

📚 References

• Dataset: Algerian Forest Fires Dataset
• Scikit-Learn: Official Documentation