Multi-label core vector machine with a zero label

作者：

Highlights：

• Multi-label core vector machine with a zero label is proposed.

• This method runs averagely 83 times faster than multi-label core vector machine.

• This method has slightly fewer support vectors than multi-label core vector machine.

• Our new algorithm is a competitive candidate for multi-label classification.

摘要

Highlights•Multi-label core vector machine with a zero label is proposed.•This method runs averagely 83 times faster than multi-label core vector machine.•This method has slightly fewer support vectors than multi-label core vector machine.•Our new algorithm is a competitive candidate for multi-label classification.

论文关键词：Multi-label classification,Support vector machine,Core vector machine,Frank–Wolfe method,Quadratic programming,Linear programming

论文评审过程：Received 13 April 2013, Revised 9 January 2014, Accepted 22 January 2014, Available online 5 February 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.01.012