anushka123jet-cloud/Algerian-Forest-Fires-Prediction-using-Regression-Models

End-to-end Machine Learning project on the Algerian Forest Fires dataset featuring data cleaning, EDA, feature engineering, visualization, regression modeling (Linear, Ridge, Lasso), model evaluation, and Pickle-based model serialization. # 🔥 Algerian Forest Fires Prediction using Machine Learning ## 📌 Project Overview This project presents a complete end-to-end Machine Learning workflow on the Algerian Forest Fires Dataset. The objective is to analyze the factors influencing forest fires and build predictive regression models capable of estimating fire-related indices based on environmental and weather conditions. The project covers every stage of a real-world Data Science pipeline, including data cleaning, exploratory data analysis (EDA), feature engineering, visualization, correlation analysis, model building, and performance evaluation. --- ## 🎯 Objectives - Clean and preprocess raw data - Perform Exploratory Data Analysis (EDA) - Discover patterns and relationships among variables - Engineer features for improved model performance - Visualize trends using statistical plots - Analyze feature correlations - Train and evaluate multiple regression models - Compare model performance using evaluation metrics --- ## 📂 Dataset Information The Algerian Forest Fires Dataset contains meteorological and environmental attributes collected from different regions of Algeria. ### Features Include: - Temperature - Relative Humidity (RH) - Wind Speed (Ws) - Rain - Fine Fuel Moisture Code (FFMC) - Duff Moisture Code (DMC) - Drought Code (DC) - Initial Spread Index (ISI) - Fire Weather Index (FWI) - Classes (Fire / Not Fire) --- ## ⚙️ Project Workflow ### 1️⃣ Data Cleaning & Preprocessing - Handled missing values - Removed inconsistencies - Corrected data types - Prepared data for analysis ### 2️⃣ Exploratory Data Analysis (EDA) - Univariate Analysis - Bivariate Analysis - Distribution Analysis - Outlier Detection ### 3️⃣ Feature Engineering - Feature selection - Data transformation - Feature preparation for modeling ### 4️⃣ Data Visualization Visualizations were created using: - Matplotlib - Seaborn Plots include: - Histograms - Box Plots - Scatter Plots - Heatmaps - Correlation Matrix - Distribu …