Ramp loss nonparallel support vector machine for pattern classification》.pdf

Knowledge-Based Systems 85 (2015) 224–233 Contents lists available at ScienceDirect Knowledge-Based Systems journal homepage: www.else /locate/knosys Ramp loss nonparallel support vector machine for pattern classification Dalian Liu a,b, Yong Shi a,c,d,e,⇑, Yingjie Tian c,d,⇑ a School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China b Department of Basic Course Teaching, Beijing Union University, Beijing 100101, China c Research Center on Fictitious Economy and Data Science, Chinese Academy of Sciences, Beijing 100190, China d Key Laboratory of Big Data Mining and Knowledge Management, Chinese Academy of Sciences, Beijing 100190, China e Collegae of Information Science and Technology, University of Nebraska at Omaha, Omaha, NE 68182, USA a r t i c l e i n f o a b s t r a c t Article history: In this paper, we propose a novel sparse and robust nonparallel hyperplane classifier, named Ramp loss Received 15 February 2015 Nonparallel Support Vector Machine (RNPSVM), for binary classification. By introducing the Ramp loss Received in revised form 16 April 2015 function and also proposing a new non-convex and non-differentiable loss function based on the e-insen- Accepted 8 May 2015 sitive loss function, RNPSVM can explicitly incorporate noise and outlier suppression in the training pro- Available online 16 May 2015 cess, has less support vectors and the increased sparsity leads to its better scaling properties. The