🚗 Multiple Linear Regression – CO2 Emissions Prediction

📌 Problem Statement

The goal of this project is to model the relationship between vehicle specifications and CO2 emissions using Multiple Linear Regression.
We aim to predict CO2EMISSIONS from features like ENGINESIZE, CYLINDERS, and FUELCONSUMPTION_COMB.

هدف پروژه مدل‌سازی رابطه‌ی بین ویژگی‌های فنی خودرو و میزان انتشار گاز CO2 با استفاده از رگرسیون خطی چندمتغیره است.

---

🧠 Solution Approach

This notebook demonstrates the full pipeline of implementing multiple linear regression using scikit-learn:

1. Data Loading
   Load the FuelConsumption.csv dataset using pandas.

2. Feature Selection
   Selected columns:
   ENGINESIZE, CYLINDERS, FUELCONSUMPTION_COMB, and CO2EMISSIONS.

3. Data Visualization
   Scatter plots to explore correlation between features and target.

4. Train-Test Split
   Dataset split randomly into training (80%) and test (20%) subsets.

5. Model Training
   A linear regression model is trained using:

   regr = LinearRegression()
   regr.fit(train_x, train_y)

``

6. Evaluation

   • Coefficients and intercept are printed.
   • R² score is computed on test data using r2_score.

---

🛠️ Technologies & Libraries

• Python 3
• Pandas – data manipulation
• NumPy – numerical operations
• Matplotlib & Pylab – plotting
• scikit-learn – LinearRegression and r2_score

---

🚀 Installation & Execution Guide

1. Install Dependencies

pip install pandas numpy matplotlib scikit-learn

2. Dataset

Place FuelConsumption.csv in the appropriate directory or update the path in the notebook accordingly.

3. Run Notebook

Use Jupyter to open and run the notebook:

jupyter notebook 02_Multiple-Linear-Regression.ipynb

---

📊 Key Results & Model Performance

• Coefficients:

  [[10.52277345, 7.57289264, 9.46033011]]
  

• R² Score: ~0.84 (reported from other version)

مدل موفق شد با دقت حدود ۸۴٪ واریانس CO2 را با استفاده از ویژگی‌ها پیش‌بینی کند.

---

🖼️ Sample Outputs

1. Raw Dataset Preview:

ENGINESIZE | CYLINDERS | FUELCONSUMPTION_COMB | CO2EMISSIONS
    2.0    |     4     |          8.5         |     196

2. Selected Features:

ENGINESIZE | CYLINDERS | FUELCONSUMPTION_COMB | CO2EMISSIONS
    2.4    |     4     |          9.6         |     221

3. Visualizations:

• ENGINESIZE vs CO2EMISSIONS
• Train/Test scatter plot
• Regression line on training data

نمودارها برای درک بهتر رابطه ویژگی‌ها و متغیر هدف استفاده شده‌اند.

---

💡 What This Project Demonstrates

• Practical understanding of multiple linear regression
• Feature selection and preprocessing
• Model training with scikit-learn
• Use of R² score for evaluation
• Professional plotting for data interpretation
• Documentation and clarity in model-building steps

---

👤 Author

Mehran Asgari 📧 imehranasgari@gmail.com 🔗 github.com/imehranasgari

---

📄 License

This project is licensed under the Apache 2.0 License.