A Comparative Study on the Implementation of Advanced Neural Networks for Predictive Modelling of Carbon Emissions in India

One of today’s most important issues is climate change, primarily due to carbon emissions. Accurate carbon emissions forecasting is crucial in developing countries like India to inform effective policy decisions and reduce environmental harm. This study addresses the gap by modeling and forecasting carbon emissions per capita based on a dataset of 172 data points from 1850 to 2022, with a special focus on emissions in India. The proposed methodology combines efficient pre-processing and data preparation with a thorough model selection process, which includes training models such as RNN, LSTM, Bidirectional LSTM, GRU and CNN-GRU. To enhance the performance of the models, hyperparameter tuning was performed. The models’ performance was evaluated using measures including mean square error (MSE), mean absolute error (MAE), and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>. The study successfully advances the use of neural networks for carbon emission prediction, which can aid in the development of climate change policy.