Agri-Eco Predict: Minimizing Carbon Intensity in Ensemble Prediction Model for Agricultural Product Price

Authors

  • R. Ragunath, R. Rathipriya

Keywords:

Agricultural Product Prices, Carbon Footprint, Ensemble Prediction Models, Green AI, Energy Consumption, Sustainability

Abstract

Computational loads of Ensemble prediction models (EPMs) are discussed in this paper with a special emphasis on their carbon footprints or carbon emissions. It is purposefully carried out to determine and reduce the carbon emissions caused by EPMs for forecasting Agricultural product prices (APP). Random Forest (RF) and Gradient Boosting (GBR), as well as combinations of both with an adaptive weighted strategy, were included in the experimental study that assessed energy consumption and carbon emissions of EPMs. Carbon emissions were significantly reduced while maintaining prediction accuracy through optimising these models on CPU and T4 GPU platforms. For instance, for optimized RF models on CPU; there was a decline in carbon emission from 3.576e-07 kgCO2e to 1.793e-07 kgCO2e, while Mean Squared Error (MSE) improved from 3.014 to 2.189 respectively. Similarly, after optimization, GBR models on GPU no longer changed their carbon footprint but changed MSE significantly. The findings indicated that it is possible to mitigate the carbon output without affecting the accuracy of predictions using hyperparameter optimization based EPM.

Downloads

Download data is not yet available.

References

Ali, A. H. (2023). Green AI for Sustainability: Leveraging Machine Learning to Drive a Circular Economy. Babylonian Journal of Artificial Intelligence, 2023, 15–16. https://doi.org/10.58496/BJAI/2023/004

Bergstra, J., & Bengio, Y. (2012). Random search for hyper-parameter optimization. Journal of Machine Learning Research.

Breiman, L. (2001). Random forests. Machine Learning. Volume 45,Page 5-32, url={https://api.semanticscholar.org/CorpusID:89141

Castellanos-Nieves, D. (2023). Improving Automated Machine-Learning Systems through Green AI, Appl. Sci. 2023, 13, 11583. https://doi.org/10.3390/app132011583

Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Ann. Statist. 29 (5) 1189 - 1232, October 2001. https://doi.org/10.1214/aos/1013203451.

Han, S., Pool, J., Tran, J., & Dally, W. (2015). Learning both weights and connections for efficient neural network. Advances in Neural Information Processing Systems.

Henderson, P., Hu, J., Romoff, J., Brunskill, E., Jurafsky, D., & Pineau, J. (2020). Towards the systematic reporting of the energy and carbon footprints of machine learning. Journal of Machine Learning Research.

Jacob, B., Kligys, S., Chen, B., Zhu, M., Tang, M., Howard, A., ... & Kalenichenko, D. (2018). Quantization and training of neural networks for efficient integer-arithmetic-only inference. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.

Khaki, S., & Wang, L. (2019). Crop yield prediction using deep neural networks. Frontiers in Plant Science.

Pantazi, X. E., Moshou, D., Alexandridis, T., Whetton, R. L., & Mouazen, A. M. (2016). Wheat yield prediction using machine learning and advanced sensing techniques. Computers and Electronics in Agriculture.

Schmidt, D., Marques, M. R., Botti, S., & Marques, M. A. (2019). Recent advances and applications of machine learning in solid-state materials science. npj Computational Materials.

Snoek, J., Larochelle, H., & Adams, R. P. (2012). Practical Bayesian optimization of machine learning algorithms. Advances in Neural Information Processing Systems.

Strubell, E., Ganesh, A., & McCallum, A. (2019). Energy and policy considerations for deep learning in NLP. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics.

Downloads

Published

09.07.2024

How to Cite

R. Ragunath. (2024). Agri-Eco Predict: Minimizing Carbon Intensity in Ensemble Prediction Model for Agricultural Product Price. International Journal of Intelligent Systems and Applications in Engineering, 12(22s), 776–781. Retrieved from https://ijisae.org/index.php/IJISAE/article/view/6554

Issue

Section

Research Article