Least squares twin multi-class classification support vector machine

作者：

Highlights：

• The multi-class classification method based on TWSVM is proposed.

• We formulate a least squares version of Twin K-class support vector machine.

• We just solve two systems of linear equations.

• Our method evaluates all the training data into a “1-v-1-v-rest” structure.

• It generates ternary output {−1, 0, +1}.

摘要

Highlights•The multi-class classification method based on TWSVM is proposed.•We formulate a least squares version of Twin K-class support vector machine.•We just solve two systems of linear equations.•Our method evaluates all the training data into a “1-v-1-v-rest” structure.•It generates ternary output {−1, 0, +1}.

论文关键词：Twin support vector machine,Least squares,Multi-class classification,Nonparallel plane,K-SVCR

论文评审过程：Received 6 July 2013, Revised 20 July 2014, Accepted 18 September 2014, Available online 30 September 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.09.020