Hyper-heuristic glowworm swarm optimised support vector machines for heart and thyroid disease classification
by G. Kiruthiga; S.Mary Vennila
International Journal of System of Systems Engineering (IJSSE), Vol. 14, No. 2, 2024

Abstract: In order to improve illness detection accuracy and reduce complexity, this study seeks to construct sophisticated machine learning (ML) classifiers and effective feature selection (FS) methods. The proposed model includes two stages: classification using Hyper-heuristic Glow Worm Swarm Optimised Support Vector Machines and FS utilising Information Gain (IG) and Spotted Hyena Optimiser (IG-SHO) (HHGWSO-SVM). By removing the irrelevant characteristics with the IG metric, the dimensionality of attributes is decreased in the IG-SHO technique. By combining the hybrid optimisation approach of HHGWSO with the SVM, the suggested HHGWSO-SVM classifier has been created. Its configuration has been improved by optimally setting the margin parameter, kernel type, and kernel parameters. The Hyperheuristic algorithm and the Glowworm Swarm Optimisation (HHGWSO) have been combined to create a method for fine-tuning SVM parameters based on accuracy and model complexity. The proposed HHGWSO-SVM model is tested in experiments on benchmark datasets to predict thyroid and heart illnesses. According to the results, the suggested categorisation model has improved precision and accuracy while reducing model complexity.

Online publication date: Fri, 01-Mar-2024

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of System of Systems Engineering (IJSSE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com