Car CO2 Emission Prediction using Linear Regression

A machine learning project that predicts CO2 emissions of cars based on engine size using linear regression analysis.

📋 Project Overview

This project analyzes the relationship between car engine size and CO2 emissions using a linear regression model. The analysis includes data visualization, model training, evaluation, and performance metrics to understand how engine specifications correlate with environmental impact.

🎯 Objectives

• Predict CO2 emissions based on car engine size
• Visualize the correlation between engine specifications and emissions
• Build and evaluate a linear regression model
• Provide insights into automotive environmental impact

📊 Dataset

The project uses the FuelConsumptionCo2.csv dataset which contains:

• Source: IBM Developer Skills Network
• Records: 1000+ vehicle entries from 2014
• Key Features:
  • ENGINESIZE: Engine displacement in liters
  • CYLINDERS: Number of engine cylinders
  • FUELCONSUMPTION_COMB: Combined fuel consumption (L/100km)
  • CO2EMISSIONS: CO2 emissions in grams per kilometer

Sample Data Structure

MODELYEAR | MAKE | MODEL | ENGINESIZE | CYLINDERS | FUELCONSUMPTION_COMB | CO2EMISSIONS
2014      | ACURA| ILX   | 2.0        | 4         | 8.5                  | 196

🛠️ Technologies Used

• Python 3.x
• Pandas: Data manipulation and analysis
• NumPy: Numerical computing
• Matplotlib: Data visualization
• Scikit-learn: Machine learning algorithms
• Requests: HTTP library for data downloading

🚀 Installation & Setup

1. Clone the repository:

   git clone https://github.com/yourusername/ML_Car_CO2_Emission_Linear_Regression_model.git
   cd ML_Car_CO2_Emission_Linear_Regression_model

2. Install required dependencies:

   pip install pandas numpy matplotlib scikit-learn requests

3. Run the model:

   python model.py

📈 Project Workflow

1. Data Collection

• Automatic download of fuel consumption dataset from IBM Cloud
• Data validation and error handling

2. Data Preprocessing

• Feature selection (Engine Size, Cylinders, Fuel Consumption, CO2 Emissions)
• Data exploration and statistical analysis

3. Exploratory Data Analysis (EDA)

• Histograms: Distribution analysis of key features
• Scatter Plots:
  • Engine Size vs CO2 Emissions
  • Fuel Consumption vs CO2 Emissions
  • Number of Cylinders vs CO2 Emissions

4. Model Development

• Data Split: 80% training, 20% testing
• Algorithm: Linear Regression
• Features: Engine Size (primary predictor)
• Target: CO2 Emissions

5. Model Training & Evaluation

• Model fitting on training data
• Performance evaluation using:
  • Mean Squared Error (MSE)
  • R² Score (Coefficient of Determination)

6. Visualization

• Training data scatter plot with regression line
• Test data predictions with fitted line
• Model performance visualization

📊 Key Findings

Model Performance Metrics

• R² Score: ~0.85-0.90 (indicating strong correlation)
• MSE: Quantifies prediction accuracy
• Correlation: Strong positive relationship between engine size and CO2 emissions

Insights

• Larger engines typically produce higher CO2 emissions
• Linear relationship exists between engine displacement and environmental impact
• Fuel consumption patterns align with emission trends

📁 Project Structure

ML_Car_CO2_Emission_Linear_Regression_model/
│
├── model.py                 # Main analysis script
├── FuelConsumption.csv     # Dataset (auto-downloaded)
├── README.md               # Project documentation
└── .git/                   # Git repository files

🔧 Usage Example

# Import required libraries
from sklearn import linear_model
import pandas as pd
import numpy as np

# Load and preprocess data
df = pd.read_csv("FuelConsumption.csv")
df = df[['ENGINESIZE', 'CYLINDERS', 'FUELCONSUMPTION_COMB', 'CO2EMISSIONS']]

# Train model
regr = linear_model.LinearRegression()
train_x = np.asanyarray(train[['ENGINESIZE']])
train_y = np.asanyarray(train[['CO2EMISSIONS']])
regr.fit(train_x, train_y)

# Make predictions
predictions = regr.predict(test_x)

📋 Requirements

pandas>=1.3.0
numpy>=1.21.0
matplotlib>=3.4.0
scikit-learn>=1.0.0
requests>=2.25.0

🤝 Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/improvement)
3. Commit changes (git commit -am 'Add new feature')
4. Push to branch (git push origin feature/improvement)
5. Create a Pull Request

👨‍💻 Author

Devansh Tomar

• GitHub: @devanshtomar
• LinkedIn: Devansh Tomar

🙏 Acknowledgments

• IBM Developer Skills Network for providing the dataset
• Scikit-learn community for machine learning tools
• Open source community for Python libraries

🔮 Future Enhancements

• Multiple linear regression with additional features
• Cross-validation for better model validation
• Feature engineering for improved accuracy
• Web application for interactive predictions
• Comparison with other regression algorithms
• Time series analysis for emission trends

---

⭐ Star this repository if you found it helpful!