A novel SVM parameter tuning method based on advanced whale optimization algorithm

The classification performance of support vector machine (SVM) algorithm is highly dependent on the careful tuning of hyper-parameters and penalty coefficient. This paper introduces a novel SVM parameter optimization method by using the advanced whale optimization algorithm (AWOA) that is an improve...

Full description

Saved in:
Bibliographic Details
Published inJournal of physics. Conference series Vol. 1237; no. 2; pp. 22140 - 22148
Main Authors Yin, Xuehao, Hou, YanDong, Yin, Jiabao, Li, Chao
Format Journal Article
LanguageEnglish
Published Bristol IOP Publishing 01.06.2019
Subjects
Algorithms
Classification
Optimization
Optimization algorithms
Parameters
Support vector machines
Tuning
Online AccessGet full text

Cover

More Information
Summary:The classification performance of support vector machine (SVM) algorithm is highly dependent on the careful tuning of hyper-parameters and penalty coefficient. This paper introduces a novel SVM parameter optimization method by using the advanced whale optimization algorithm (AWOA) that is an improved whale of algorithm (WOA) with external archiving strategy. A new framework for SVM parameter optimization based on AWOA is built. To demonstrate the performance of our proposed method, six typical data sets are chosen to evaluate the effect of SVM classification problem. Experimental results show that the higher accuracy and better convergence can be achieved by AWOA compared with other three usual parameter optimization methods (WOA, PSO, and DE).
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/1237/2/022140