Home > Papers

 
 
A New Online Algorithm for Multiclass of SVM
zhang leilei 1 *,Sheng Baohuai 2
1.Mathmatical department of NingBo university
2.Department of Mathematics, Shaoxing University
*Correspondence author
#Submitted by
Subject:
Funding: 国家自然科学基金（No.10871226）
Opened online:25 March 2009
Accepted by: none
Citation: zhang leilei,Sheng Baohuai.A New Online Algorithm for Multiclass of SVM[OL]. [25 March 2009] http://en.paper.edu.cn/en_releasepaper/content/30712
 
 
In the present paper we provide a new algorithm for muticlass vector machines. Our starting point is a generalized notion of the maximal margin to multiclass problems. Using this notion we cast the muticlass problem as a binary classification task every time. So we combine a new classification function with the approximate maximal margin algorithm which is devised for binary classification.Our algorithm needs O( (p−1) α2γ2 ) corrections every time to separate the data with p-norm margin larger than l(1 − α)γ, with γ being the p-norm margin of the data.
Keywords:Multiclass;Support vector Machines;On-line learning;Classification;Maximal
 
 
 

For this paper
  • ● PDF (0B)
  • ● Revision 0 　　
  • ● Print this paper
  • ● Recommend this paper to a friend
  • ● Add to my favorite list

    Saved Papers

    Please enter a name for this paper to be shown in your personalized Saved Papers list

Tags
Add yours

Related Papers

Statistics
PDF Downloaded 259
Bookmarked 0
Recommend 5
Comments Array
Submit your papers