Dimension reduction based on a penalized kernel support vector machine model
作者：
Highlights：
• Penalized Kernel Support Vector machines (PKSVM) model combined with Support Vector Information Criterion (SVMIC) is proposed.
• We reformulate the PKSVM model as a linear-in-the-parameters problem.
• We derive a PKSVM+SVMIC algorithm which is easy to implement and computational efficient.
• Both 10-fold Cross Validation and Support Vector Information Criterion are utilized to optimize the model parameters.
• The experiment reveals that our developed models get better performance.
摘要
•Penalized Kernel Support Vector machines (PKSVM) model combined with Support Vector Information Criterion (SVMIC) is proposed.•We reformulate the PKSVM model as a linear-in-the-parameters problem.•We derive a PKSVM+SVMIC algorithm which is easy to implement and computational efficient.•Both 10-fold Cross Validation and Support Vector Information Criterion are utilized to optimize the model parameters.•The experiment reveals that our developed models get better performance.
论文关键词：Dimension reduction,SVM,PKSVM,SVMIC
论文评审过程：Received 27 August 2016, Revised 19 July 2017, Accepted 30 September 2017, Available online 2 October 2017, Version of Record 13 November 2017.
论文官网地址：https://doi.org/10.1016/j.knosys.2017.09.041
