Robust twin support vector regression via second-order cone programming

作者:

Highlights:

• Novel robust approach for twin SVR using second-order cone programming.

• A geometrically grounded method based on the concept of ellipsoids.

• Superior performance is achieved in experiments on benchmark datasets.

摘要

•Novel robust approach for twin SVR using second-order cone programming.•A geometrically grounded method based on the concept of ellipsoids.•Superior performance is achieved in experiments on benchmark datasets.

论文关键词:Support vector regression,Twin support vector machines,Second-order cone programming,Robust optimization

论文评审过程:Received 9 September 2017, Revised 1 April 2018, Accepted 2 April 2018, Available online 3 April 2018, Version of Record 12 May 2018.

论文官网地址:https://doi.org/10.1016/j.knosys.2018.04.005