New Hermite orthogonal polynomial kernel and combined kernels in Support Vector Machine classifier
作者：
Highlights：
• A new kernel function has been proposed by using Hermite orthogonal polynomial.
• New combined kernels resulted by combining Hermite and common kernels have been proposed.
• The proposed methods have been compared with three common kernels on real and synthetic data sets.
• Hermite kernel function has the lowest required time for classification.
• The proposed methods yield better error rate and they have the best support vectors reduction significantly.
摘要
•A new kernel function has been proposed by using Hermite orthogonal polynomial.•New combined kernels resulted by combining Hermite and common kernels have been proposed.•The proposed methods have been compared with three common kernels on real and synthetic data sets.•Hermite kernel function has the lowest required time for classification.•The proposed methods yield better error rate and they have the best support vectors reduction significantly.
论文关键词：Support Vector Machine (SVM),Kernel function,Hermite orthogonal polynomial kernel,Combined kernel
论文评审过程：Received 4 July 2014, Revised 9 June 2016, Accepted 3 July 2016, Available online 7 July 2016, Version of Record 26 July 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.07.004
