Keywords
Biogeography- Based optimization (BBO)
Explainable AI
SHAP
LIME
How to Cite
Abstract
In this study, the Biogeography-Based Optimization (BBO) algorithm has been effectively utilized to predict carbon dioxide (CO2) emissions. BBO is a nature-inspired optimization method used to analyze complex relationships and determine significant features in a dataset. The focus of the study is to accurately predict the "share_global_coal_co2" parameter, and for this purpose, the BBO algorithm has been used to identify the most effective 20 features. The analyses revealed that the Gradient Boosting algorithm provided the lowest Mean Squared Error (MSE) value (0.347408), indicating that the model could make predictions closer to the real data. Additionally, with the use of interpretable artificial intelligence models such as SHAP and LIME, it was determined that the model's predictions became more understandable, and the effects of specific features on the predictions were clarified. The obtained results provide important guidance for environmental policymakers and energy experts. The effectiveness of the BBO algorithm in predicting CO2 emissions can contribute to making more knowledge-based and data-driven decisions in environmental analysis and policymaking processes. This study highlights the importance of artificial intelligence and optimization techniques in achieving sustainability goals and helps in developing more effective strategies in environmental management.
References
Ali, S., T. Abuhmed, S. El-Sappagh, K. Muhammad, J. M. Alonso- Moral, et al., 2023 Explainable artificial intelligence (xai): What we know and what is left to attain trustworthy artificial intelligence. Information fusion 99: 101805.
Bentéjac, C., A. Csörg˝o, and G. Martínez-Muñoz, 2021 A comparative analysis of gradient boosting algorithms. Artificial Intelligence Review 54: 1937–1967.
Chen, T. and C. Guestrin, 2016 Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, pp. 785–794.
Das, A. and P. Rad, 2020 Opportunities and challenges in explainable artificial intelligence (xai): A survey. arXiv preprint arXiv:2006.11371 .
Delanoë, P., D. Tchuente, and G. Colin, 2023 Method and evaluations of the effective gain of artificial intelligence models for reducing co2 emissions. Journal of Environmental Management 331: 117261.
Gunn, S. R., 1998 Support vector machines for classification and regression. ISIS technical report 14: 5–16.
Heo, S., J. Ko, S. Kim, C. Jeong, S. Hwangbo, et al., 2022 Explainable ai-driven net-zero carbon roadmap for petrochemical industry considering stochastic scenarios of remotely sensed offshore wind energy. Journal of Cleaner Production 379: 134793.
Kramer, O. and O. Kramer, 2013 K-nearest neighbors. In Dimensionality reduction with unsupervised nearest neighbors, pp. 13–23, Springer.
Liaw, A. and M. Wiener, 2002 Classification and regression by randomforest. R news 2: 18–22.
Qerimi, Q. and B. S. Sergi, 2022 The case for global regulation of carbon capture and storage and artificial intelligence for climate change. International Journal of Greenhouse Gas Control 120: 103757.
Stef, N., H. Ba¸sa˘gao˘ glu, D. Chakraborty, and S. B. Jabeur, 2023 Does institutional quality affect co2 emissions? evidence from explainable artificial intelligence models. Energy Economics 124: 106822.
Thanh, H. V., A. Zamanyad, M. Safaei-Farouji, U. Ashraf, and mZ. Hemeng, 2022 Application of hybrid artificial intelligent models to predict deliverability of underground natural gas storage sites. Renewable Energy 200: 169–184.
Yan, H., J. Zhang, N. Zhou, and M. Li, 2020 Application of hybrid artificial intelligence model to predict coal strength alteration during co2 geological sequestration in coal seams. Science of the otal environment 711: 135029.
Zhang, Y., C. Zhu, and Q.Wang, 2020 Lightgbm-based model for metro passenger volume forecasting. IET Intelligent Transport Systems 14: 1815–1823.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.