DarshAwasthi2005/Algerian-_Forestfire_Model

🔥 Algerian Forest Fires – Machine Learning Regression Analysis 📌 Overview This project performs predictive modeling on the Algerian Forest Fires Dataset to estimate fire weather index–related outcomes using multiple regression techniques # Algerian-_Forestfire_Model 🔥 Algerian Forest Fires – Machine Learning Regression Analysis 📌 Overview This project performs predictive modeling on the Algerian Forest Fires Dataset to estimate fire weather index–related outcomes using multiple regression techniques. The objective is to: Perform structured data preprocessing Analyze feature correlation Apply feature encoding and scaling Train multiple regression models Evaluate and compare model performance 📂 Dataset Information Dataset: Algerian Forest Fires Dataset Source: Public wildfire dataset containing meteorological and fire index attributes Features Include: Temperature Relative Humidity (RH) Wind Speed (Ws) Rain FFMC DMC DC ISI BUI FWI Region / Class (Categorical) 🛠 Tech Stack Python 3.x NumPy Pandas Matplotlib Seaborn Scikit-Learn 📊 Project Workflow 1️⃣ Data Preprocessing Loaded cleaned dataset Removed inconsistencies Converted categorical features using Label Encoding Performed correlation analysis Removed highly correlated features using custom threshold function 2️⃣ Train-Test Split train_test_split(X, Y, test_size=0.2, random_state=42) 80% training data 20% testing data Random state fixed for reproducibility 3️⃣ Feature Scaling Standard scaling applied to normalize feature distributions. Improves regression stability Prevents bias due to feature magnitude differences Visualization: Boxplot before scaling Boxplot after scaling 🤖 Models Implemented The following regression models were trained and evaluated: ✔ Linear Regression ✔ Lasso Regression ✔ Ridge Regression ✔ ElasticNet Regression 📈 Evaluation Metrics Models were evaluated using: Mean Absolute Error (MAE) R² Score 🔹 Best Observed Performance Mean Absolute Error: 2.1428 R² Score: 0.8807 Interpretation MAE ≈ 2.14 → Low average prediction error R² ≈ 0.88 → Model explains 88% of variance Indicates strong predictive capability 📉 Visualization Correlation Heatmap Feature Scaling Compa …