Performance Testing of the Backpropagation Method in Predicting Carbon Dioxide Emissions Based on Root Mean Square Error and Mean Absolute Error Results

Authors : Christofer Satria; I Putu Hariyadi; Lilik Widyawati; Anthony Anggrawan; Peter Wijaya Sugijanto et al.
article cite 0 Year 2025
source:
Abstract

Carbon dioxide (CO2) emissions are a type of gas that is highly detrimental to the environment, accounting for up to 80% of total global greenhouse gas emissions. Therefore, it is not surprising that researchers recognize the need to conduct immediate research on CO2 emissions. Besides that, making predictions without using intelligent tools often results in wrong decisions. In the meantime, Backpropagation has gained widespread use for prediction as one of the algorithms that exhibits intelligence. In turn, research on CO2 is a significant research challenge today. That is why this research aims to predict CO2 emissions using the Backpropagation method, based on the Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The findings of this study revealed that the CO2 gas emission prediction had an RMSE of 0.063 and an MAE of 0.032. These findings indicate that the proposed model in this study can predict well, with no significant signs of overfitting, and yields a prediction result with low error.


Concepts :
Air Quality Monitoring and Forecasting
Data Mining and Machine Learning Applications
Edcuational Technology Systems
article cite 0 Year 2025 source
SDGs
Responsible consumption and production
Citations by Year
YearCount
2025 0