The system will be unavailable due to maintenance on Thursday July 19 from 7:00-8:30 am ET.

Show simple item record

dc.contributorVirginia Tech
dc.contributor.authorWang, Z. B.
dc.contributor.authorFang, S. C.
dc.contributor.authorGao, D. Y.
dc.contributor.authorXing, W. X.
dc.identifier.citationWang, Z. B.; Fang, S. C.; Gao, D. Y.; Xing, W. X., "Global extremal conditions for multi-integer quadratic programming," J. Industrial and Management Optimization 4(2), 213-225, (2008); DOI: 10.3934/jimo.2008.4.213
dc.description.abstractSupport vector machine (SVM) is a very popular method for binary data classification in data mining ( machine learning). Since the objective function of the unconstrained SVM model is a non-smooth function, a lot of good optimal algorithms can't be used to find the solution. In order to overcome this model's non-smooth property, Lee and Mangasarian proposed smooth support vector machine (SSVM) in 2001. Later, Yuan et al. proposed the polynomial smooth support vector machine (PSSVM) in 2005. In this paper, a three-order spline function is used to smooth the objective function and a three-order spline smooth support vector machine model (TSSVM) is obtained. By analyzing the performance of the smooth function, the smooth precision has been improved obviously. Moreover, BFGS and Newton-Armijo algorithms are used to solve the TSSVM model. Our experimental results prove that the TSSVM model has better classification performance than other competitive baselines.
dc.description.sponsorshipTsinghua Basic Research Foundation # 052201070
dc.description.sponsorshipUS NSF Grant # DMI-0553310, CCF-0514768
dc.publisherAmerican Institute of Mathematical Sciences
dc.subjectquadratic programming
dc.subjectdata mining
dc.subjectsupport vector machine
dc.subjectconstrained variational-inequalities
dc.subjectunconstrained optimization
dc.subjectglobal optimization
dc.subjectperfect duality
dc.subjectengineering, multidisciplinary
dc.subjectoperations research & management
dc.subjectmathematics, interdisciplinary applications
dc.titleGlobal extremal conditions for multi-integer quadratic programming
dc.title.serialJournal of Industrial and Management Optimization

Files in this item


This item appears in the following Collection(s)

Show simple item record