Revealing the impact of the combination of parameters on SVM performance in COVID-19 classification

Prasetiyowati, Suryani, Harini, Sri ORCID: https://orcid.org/0000-0001-9664-027X, Fadila, Juniardi Nur ORCID: https://orcid.org/0000-0002-5965-9243 and Fahlena, Hilda (2024) Revealing the impact of the combination of parameters on SVM performance in COVID-19 classification. International Journal on Information and Communication Technology (IJoICT), 10 (1). pp. 127-140. ISSN 2356-5462

Text 20196.pdf - Published Version Download (513kB)

Abstract

SVM is a method that has advantages in classification, but there are still obstacles in selecting optimal parameters. This research investigates the effect of parameter variations on SVM classification performance on the COVID-19 dataset, using linear, RBF, Sigmoid and polynomial kernels. Experiments were carried out to evaluate the combination of kernel parameter and regulations coefficient in each kernel against performance metrics such as accuracy, F1-Score, recall, and precision. In this study, the highest accuracy is 77.57% with an F1-Score of 76.67% when the parameter kernel and regulations coefficient are 0.75 and 0.75, respectively. The result is indicating an optimal balance between precision and recall. The other hand, the performance stability of polynomial kernel provides an optimal approach for analysis and prediction in classifying COVID-19 dataset, with more controlled fluctuations than other kernels. The interaction between the C and Gamma parameters shows that a Gamma value of 0.75 consistently provides good results, while adjusting the C parameter shows more controlled performance variations. This confirms that appropriate Gamma parameter settings are key in improving the accuracy and consistency of SVM model predictions in this case. This research provides insight into the potential of implementing SVM with polynomial kernels for COVID-19 classification, in analysis and prediction on complex biomedical datasets

Item Type:	Journal Article
Keywords:	Support Vector Machine; Kernel; Parameter; Performance; Covid-19
Subjects:	01 MATHEMATICAL SCIENCES > 0103 Numerical and Computational mathematics
Divisions:	Faculty of Mathematics and Sciences > Department of Mathematics
Depositing User:	Prof SRI HARINI
Date Deposited:	23 Aug 2024 09:37

Downloads

Origin of downloads

Actions (login required)

View Item