A generalized Gilbert algorithm and an improved MIES for one-class support vector machine

作者：

Highlights：

• The primal maximum margin problem of OCSVM is equivalent to a nearest point problem.

• A generalized Gilbert (GG) algorithm is proposed to solve the nearest point problem.

• An improved MIES is developed for the Gaussian kernel parameter selection.

• The GG algorithm is computationally more efficient than the SMO algorithm.

摘要

•The primal maximum margin problem of OCSVM is equivalent to a nearest point problem.•A generalized Gilbert (GG) algorithm is proposed to solve the nearest point problem.•An improved MIES is developed for the Gaussian kernel parameter selection.•The GG algorithm is computationally more efficient than the SMO algorithm.

论文关键词：One-class support vector machine,Generalized Gilbert algorithm,Reduced convex hull,Gaussian kernel,Parameter selection

论文评审过程：Received 5 May 2015, Revised 13 September 2015, Accepted 16 September 2015, Available online 26 September 2015, Version of Record 8 November 2015.

论文官网地址：https://doi.org/10.1016/j.knosys.2015.09.016