A unified approach to support vector machines Shilton, Alistair and Palaniswami, Marimuthu 2008, A unified approach to support vector machines. In Verma, Brijesh and Blumenstein, Michael (ed), Pattern Recognition Technologies and Applications: Recent Advances, IGI Global, Hershey, Pa., pp.299-324, doi: 10.4018/978-1-59904-807-9.ch014. Attached Files Name Description MIMEType Size Downloads Title A unified approach to support vector machines Author(s) Shilton, Alistair orcid.org/0000-0002-0849-3271 Palaniswami, Marimuthu Title of book Pattern Recognition Technologies and Applications: Recent Advances Editor(s) Verma, Brijesh Blumenstein, Michael Publication date 2008 Chapter number 14 Total chapters 17 Start page 299 End page 324 Total pages 26 Publisher IGI Global Place of Publication Hershey, Pa. Summary This chapter presents a unified introduction to support vector machine (SVM) methods for binary classification, one-class classification, and regression. The SVM method for binary classification (binary SVC) is introduced first, and then extended to encompass one-class classification (clustering). Next, using the regularized risk approach as a motivation, the SVM method for regression (SVR) is described. These methods are then combined to obtain a single unified SVM formulation that encompasses binary classification, one-class classification, and regression (as well as some extensions of these), and the dual formulation of this unified model is derived. A mechanical analogy for binary and one-class SVCs is given to give an intuitive explanation of the operation of these two formulations. Finally, the unified SVM is extended to implement general cost functions, and an application of SVM classifiers to the problem of spam e-mail detection is considered. ISBN 1599048078 9781599048079 Language eng DOI 10.4018/978-1-59904-807-9.ch014 Indigenous content off HERDC Research category BN.1 Other book chapter, or book chapter not attributed to Deakin Copyright notice ©2008, IGI Global Persistent URL http://hdl.handle.net/10536/DRO/DU:30125477 Document type: Book Chapter Collections: Faculty of Science, Engineering and Built Environment Centre for Pattern Recognition and Data Analytics (PRADA) Related Links Link Description Connect to published version Go to link with your DU access privileges     Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved. Versions Version Filter Type Thu, 11 Jul 2019, 13:21:13 EST Fri, 12 Jul 2019, 05:02:11 EST Fri, 12 Jul 2019, 06:03:54 EST Wed, 11 Nov 2020, 07:42:08 EST Wed, 11 Nov 2020, 07:57:29 EST Wed, 11 Nov 2020, 11:16:26 EST Filtered Full Citation counts:   Cited 0 times in TR Web of Science Cited 2 times in Scopus Search Google Scholar Access Statistics: 26 Abstract Views, 0 File Downloads  -  Detailed Statistics Created: Thu, 11 Jul 2019, 13:21:13 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.