Optimizing Initial Guesses for Nonlinear System Solvers Using Machine Learning: A Comparative Study of Classification Algorithms DOI Open Access

Iven Aabaah,

Japheth Kodua Wiredu,

Bakaweri Emmanuel Batowise

et al.

International Journal of Computer Science and Engineering, Journal Year: 2024, Volume and Issue: 11(12), P. 7 - 15

Published: Dec. 30, 2024

This paper focuses on the problem of improving initial guesses provided to solvers nonlinear systems in terms enhancing both convergence efficiency and reliability.A novel approach for constructing confidence models is proposed based a Logistic Regression, Support Vector Machines (SVM), Random Forests, K-Nearest Neighbors (KNN) classification schemes.Experimental evaluation across diverse highlights Forests as most effective model with an average accuracy 81.69%, precisionof 83.23%, recallof 82.16%, F1 score 82.69% highest AUC equal 0.90.Backed up by broad metrics, above research inquiries mark ideal potential machine learning revolutionize data processing increasing solver adaptability, patterns economizing computations scientific engineering modalities.

Language: Английский

Optimizing Initial Guesses for Nonlinear System Solvers Using Machine Learning: A Comparative Study of Classification Algorithms DOI Open Access

Iven Aabaah,

Japheth Kodua Wiredu,

Bakaweri Emmanuel Batowise

et al.

International Journal of Computer Science and Engineering, Journal Year: 2024, Volume and Issue: 11(12), P. 7 - 15

Published: Dec. 30, 2024

This paper focuses on the problem of improving initial guesses provided to solvers nonlinear systems in terms enhancing both convergence efficiency and reliability.A novel approach for constructing confidence models is proposed based a Logistic Regression, Support Vector Machines (SVM), Random Forests, K-Nearest Neighbors (KNN) classification schemes.Experimental evaluation across diverse highlights Forests as most effective model with an average accuracy 81.69%, precisionof 83.23%, recallof 82.16%, F1 score 82.69% highest AUC equal 0.90.Backed up by broad metrics, above research inquiries mark ideal potential machine learning revolutionize data processing increasing solver adaptability, patterns economizing computations scientific engineering modalities.

Language: Английский

Citations

0