Evolutionary wrapper approaches for training set selection as preprocessing mechanism for support vector machines: Experimental evaluation and support vector analysis
详细信息 查看全文
- 作者：Nele Verbiesta ; Nele.Verbiest@UGent.be" class="auth_mail" title="E-mail the corresponding author ; Joaquí ; n Derracb ; Chris Cornelisa ; c ; Salvador Garcí ; ac ; d ; Francisco Herrerac
- 关键词：Support vector machines ; Training set selection ; Data reduction
- 刊名：Applied Soft Computing
- 出版年：2016
- 出版时间：January 2016
- 年：2016
- 卷：38
- 期：Complete
- 页码：10-22
- 全文大小：1465 K
文摘
- •
Support vector machines (SVMs) are popular and accurate classifiers.
- •
We study if SVMs can be further improved using training set selection (TSS).
- •
We adjust wrapper TSS techniques for SVMs.
- •
Experimental evaluation shows that filter TSS techniques cannot improve the accuracy of SVMs.
- •
Experimental evaluation shows that evolutionary based wrapper TSS techniques significantly improve SVMs.