KEERTHANANAMULAMETU/Algerian-Forest-Fires-Regression

# **Algerian Forest Fires Prediction** This project predicts the **Fire Weather Index (FWI)** using historical weather and FWI component data. The prediction of FWI is essential for forest fire risk assessment and mitigation strategies. --- ## **Table of Contents** 1. Introduction 2. Problem Statement 3. Objectives 4. Dataset Description 5. Methodology 6. Models and Evaluation 7. Results 8. Installation 9. Usage 10. Future Work --- ## **1. Introduction** Forest fires have devastating impacts on ecosystems, economies, and human lives. The **Fire Weather Index (FWI)** is widely used to estimate the risk of forest fires based on meteorological conditions and forest dryness. Accurate FWI predictions are essential for timely interventions and effective resource management. This project uses machine learning regression models to predict FWI based on weather and FWI component data for two Algerian regions: **Bejaia** and **Sidi Bel-abbes**. --- ## **2. Problem Statement** Managing forest fires requires accurate and timely prediction of fire risk. Existing systems may lack precision or fail to generalize across regions. This project addresses these gaps by developing a machine learning model capable of accurately predicting FWI using historical data. --- ## **3. Objectives** 1. Develop regression models to predict the Fire Weather Index (FWI). 2. Identify key features influencing FWI to support decision-making. 3. Create a scalable and easily deployable system for FWI prediction. --- ## **4. Dataset Description** - **Source**: Weather and FWI component data for the Bejaia and Sidi Bel-abbes regions in Algeria (June–September 2012). - **Features**: - **Weather Variables**: Temperature, Humidity, Wind Speed, Rain. - **FWI Components**: FFMC, DMC, DC, ISI, BUI. - **Target Variable**: Fire Weather Index (FWI). - **Size**: 244 observations with 11 variables. --- ## **5. Methodology** ### **1. Data Preprocessing** - Handled missing values by replacing them wit …