Towards greener futures: SVR-based CO₂ prediction model boosted by SCMSSA algorithm.

Area of Science:

• Environmental Science
• Computer Science
• Optimization Algorithms

Background:

• Accurate CO2 prediction is crucial for environmental sustainability and climate change mitigation.
• Existing optimization algorithms often face challenges with convergence speed and accuracy in complex prediction tasks.
• Support Vector Regression (SVR) is a powerful tool for prediction but can be enhanced for improved performance.

Purpose of the Study:

• To introduce an enhanced Sine cosine perturbation with Chaotic perturbation and Mirror imaging strategy-based Salp Swarm Algorithm (SCMSSA).
• To evaluate the performance of SCMSSA against other optimization algorithms using standard test functions.
• To assess the efficacy of SCMSSA in improving Support Vector Regression (SVR) models for CO2 prediction.

Main Methods:

• Development and implementation of the enhanced Sine cosine perturbation with Chaotic perturbation and Mirror imaging strategy-based Salp Swarm Algorithm (SCMSSA).
• Performance evaluation of SCMSSA using six benchmark test functions.
• Integration of SCMSSA with Support Vector Regression (SVR) for CO2 prediction and comparison with existing models.

Main Results:

• The SCMSSA algorithm demonstrated enhanced convergence speed and accuracy compared to other optimization algorithms.
• The SVR-SCMSSA hybrid model achieved 95% accuracy in CO2 prediction, outperforming standard SVR and other hybrid models.
• Feature importance analysis identified fossil fuel, Biomass, and Wood as significant contributors to CO2 emissions.

Conclusions:

• The SCMSSA algorithm offers superior precision and robustness for complex optimization problems.
• The SVR-SCMSSA hybrid model provides a highly accurate and reliable approach for CO2 prediction.
• Findings support the adoption of SVR-SCMSSA for environmental sustainability and climate change mitigation efforts.