The escalating emergence of environmental issues, including the greenhouse effect and coral bleaching, has raised global awareness of the significance of sustainable development and preservation of the earth's resources. Although reducing emissions is essential to mitigate the adverse effects of greenhouse gases, it remains challenging to detect without specialized equipment. This constraint is particularly burdensome for small organizations and individual groups due to the high associated costs. Therefore, this study proposes using machine learning algorithms and common vehicle attributes to predict greenhouse gas emissions accurately. Specifically, the research employs the random forest algorithm, incorporating vehicle power parameters to predict carbon dioxide emissions. The study employs Mean Square Error (MSE), Root Mean Square Error (RMSE), and R-squared metrics to analyze the model's effectiveness, accuracy, and feasibility in predicting greenhouse gas emissions. This approach will enable small groups to participate in environmental protection efforts, democratizing the process for all who desire to safeguard the environment.