Innovation, Crude Oil Prices, Fossil Fuel Energy Consumption and Climate Sustainability in Egypt: Using the Gradient Boosting Machine Learning Algorithm

The present paper analyses the predictions of yearly carbon emissions and the impacts of innovation, crude oil prices, fossil fuel energy consumption, urbanisation, and economic growth on climate sustainability in Egypt from 1990 to 2024 using a gradient-boosting machine learning algorithm. The models' performance was assessed using root mean square error (RMSE), mean absolute error (MAE), R-squared, accuracy, precision, recall, F1 score, area under the curve (ROC AUC), and confusion matrix accuracy. The findings demonstrated that a gradient-boosting algorithm attains nearly flawless performance across all assessment metrics for carbon emission prediction. Moreover, the analysis revealed that fossil fuel demand is dominant in explaining carbon emissions in Egypt. It means fossil fuel consumption is the most influential factor at 22.3%, followed by economic growth at 20.2%, innovation at 15.6%, crude oil prices at 14.8%, renewable energy consumption at 14.2%, and finally urban population at 12.9%. The document offers substantial implications for policymakers and academics in mitigating CO? emissions. The findings suggest that it is necessary to implement a comprehensive policy package that emphasises demand management, enhances energy efficiency programs, accelerates clean energy deployment, and promotes green innovation in pursuit of a more sustainable energy mix in Egypt in the long term.