Support Vector Data Description by using hyper-ellipse instead of hyper-sphere

Author

Forghani, Y. ; Yazdi, Hadi Sadoghi ; Effati, Sohrab ; Tabrizi, Reza Sigari

Author_Institution

Comput. Dept., Azad Univ., Mashhad, Iran

fYear

2011

fDate

13-14 Oct. 2011

Firstpage

22

Lastpage

27

Abstract

Support Vector Data Description (SVDD) describes data by using a hyper-sphere. In this paper, we propose an extended SVDD (ESVDD) which describes data by using a hyper-ellipse. Clearly, ESVDD can describe data better than SVDD in the input space. Both hyper-sphere and hyper-ellipse are very rigid for data description. The kernel ESVDD which will be proposed in this paper and the kernel SVDD enhance the ability of ESVDD and SVDD for data description, respectively. The formulation of SVDD/ESVDD contains a penalty term C which controls the tradeoff between the volume of hyper-sphere/hyper-ellipse and the training errors. We show that the ESVDD can control this tradeoff better than the SVDD.

Keywords

pattern classification; support vector machines; ESVDD; extended SVDD; hyper-ellipse; hyper-sphere; input space; kernel ESVDD; one-class classification method; penalty term C; support vector data description; support vector machine; training errors; Computers; Educational institutions; Kernel; Noise measurement; Support vector machines; Training; Vectors; Data description; ESVDD; Hyper-ellipse; Kernel;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer and Knowledge Engineering (ICCKE), 2011 1st International eConference on

Conference_Location

Mashhad

Print_ISBN

978-1-4673-5712-8

Type

conf

DOI

10.1109/ICCKE.2011.6413318

Filename

6413318