This project focuses on forecasting CO₂ emissions using a combination of time-series analysis and machine learning models. The primary goal is to predict future CO₂ emissions based on historical data and identify trends to support decision-making for environmental policies and sustainable development.
The dataset includes historical CO₂ emission data with attributes such as:
-
Year
-
Country/Region
-
Emission levels
-
Platform: Kaggle Notebooks
-
Programming Language: Python
-
Libraries: Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn, Statsmodels, FBProphet
-
Modeling Frameworks: ARIMA, Prophet, Regression Models
Accessing the Notebook
Visit the Kaggle Notebook.
Fork the notebook to your Kaggle account.
Run the notebook to preprocess the dataset, train models, and visualize forecasts.
-
Data Preprocessing: Cleaning and scaling the dataset.
-
Exploratory Data Analysis: Visualizing trends and relationships.
-
Model Training: Implementing ARIMA, Prophet, and regression models.
-
Evaluation: Comparing model results and selecting the best-performing model.
-
Forecasting: Generating future predictions and analyzing trends.
-
We can clearly see that double exponential smoothing model is performing best among all other models
-
Try playing with model parameters, try to understand them and tune them and you'll see improvement in your model.
Incorporate more advanced machine learning models such as XGBoost or LSTM.
Enhance dataset by integrating external factors like economic and environmental indicators.
Feel free to contribute to this project by submitting pull requests or opening issues for suggestions and improvements.
This project is licensed under the MIT License. See the LICENSE file for details.