Exploring Optimization Strategies for Support Vector Machine -Based Half Cell Potential Prediction

Abstract

Purpose This study aims to evaluate the predictive performance of Support Vector Machine (SVM) models in estimating HCP values based on input parameters, employing Bayesian Optimization, Grid Search, and Random Search optimization techniques. Study Design/Methodology Using a dataset containing 1134 rows and six columns, Principal Component Analysis (PCA) is utilized to reduce dimensionality while preserving 95% of the explained variance. Input parameters such as temperature, age, relative humidity, and X and Y lengths are considered, with HCP as the output parameter. Various performance metrics including RMSE, R-squared, MSE, MAE, prediction speed, and training time are analyzed to evaluate model effectiveness. Findings The study reveals variations in predictive performance across different optimization methods, with RMSE values ranging from 18.365 to 30.205 and R-squared values spanning from 0.88 to 0.96. Additionally, differences in training times, prediction speeds, and model complexities are observed, highlighting the trade-offs between model accuracy and computational efficiency. Originality This study contributes to the understanding of SVM model efficacy in HCP prediction, emphasizing the importance of optimization techniques, model complexity, and dimensionality reduction methods such as PCA. The comprehensive evaluation of performance metrics across different optimization methods provides valuable insights for researchers and practitioners in the field.