基于类间相似方向数的二叉树支持向量机

详细信息    查看全文 | 推荐本文 |

英文篇名：Binary Tree Support Vector Machine Based on Number of Inter-class Similarity Direction 作者：王建建 ; 何枫 英文作者：Wang Jianjian;He Feng;Donlinks School of Economics and Management, University of Science & Technology Beijing; 关键词：二叉树 ; 支持向量机 ; 类间相似方向数 ; 多分类 英文关键词：binary tree;;support vector machine;;inter-class similar direction number;;multi-classification 中文刊名：TJJC 英文刊名：Statistics & Decision 机构：北京科技大学东凌经济管理学院; 出版日期：2018-02-11 16:45 出版单位：统计与决策 年：2018 期：v.34;No.496 基金：国家自然科学基金资助项目(71272160;71673022);; 教育部科学技术战略研究资助项目(2015KJW02);教育部科技技术委员会战略研究资助项目(KJW-A-1410;GX2015-1008(Y));; 中央高校基本科研业务费专项资金资助项目(FRF-BR-16-002A) 语种：中文; 页：TJJC201804004 页数：5 CN：04 ISSN：42-1009/C 分类号：17-21

摘要

二叉树支持向量机是多分类的方法之一,其分类效果与二叉树结构有很大的关系。文章针对二叉树结构的分类顺序对分类精度影响的问题,改进了一种类间相似方向,并结合距离,提出了类间相似方向数作为生成偏二叉树支持向量机的多分类方法,弥补了距离不能很好地反映类分离度的缺陷。采用生成二叉树结构的方法进行数据实验分析,表明该方法具有较高的分类精度和分类效率。
        Binary tree support vector machine is one of the multi-classification methods, and its classification effect has much to do with the binary tree structure. Aiming at the influence of classification order of binary tree structure on classification accuracy, this paper improves a kind of inter-class similarity direction, and combines with distance to propose a multi-classification method which takes inter-class similarity direction number as generated partial binary tree support vector machine so as to overcome the defect of the distance failure to reflect the classification separation degree. Finally the paper adopts the method of generating binary tree structure to conduct a data experiment analysis, which indicates that the proposed method has higher classification accuracy and efficiency.

引文

[1]Vapnik V N.The Nature of Statistical Learning Theory[M]New York:Springer Verlag,1995.
    [2]Madzarov G,Gjorgjevikj D,Chorbev I.A Multi-class SVM Classifier Utilizing Binary Decision Tree[J].Informatica,2009,33(2).
    [3]Hill S I,Doucet A.A Framework for Kernel-Based Multi-Category Classification[J].Artif.Intell.Res.(JAIR),2007,(30).
    [4]He X,Wang Z,Jin C,et al.A Simplified Multi-Class Support Vector Machine With Reduced Dual Optimization[J].Pattern Recognition Letters,2012,33(1).
    [5]Martínez J,Iglesias C,Matías J M,et al.Solving the Slate Tile Classification Problem Using a DAGSVM Multiclassification Algorithm Based on SVM binary Classifiers With A One-Versus-All Approach[J].Applied Mathematics and Computation,2014,(230).
    [6]Du P,Tan K,Xing X.A Novel Binary Tree Support Vector Machine for Hyperspectral Remote Sensing Image Classification[J].Optics Communications,2012,285(13).
    [7]Cheong S,Oh S H,Lee S Y.Support Vector Machines With Binary Tree Architecture for Multi-Class Classification[J].Neural Information Processing-Letters and Reviews,2004,2(3).
    [8]孟媛媛,刘希玉.一种新的基于二叉树的SVM多类分类分法[J].计算机应用,2005,25(11).
    [9]唐发明,王仲东,陈绵云.支持向量机多类分类算法研究[J].控制与决策,2005,20(7).
    [10]唐发明,王仲东,陈绵云.一种新的二叉树多类支持向量机算法[J].计算机工程与应用,2005,20(7).
    [11]汪洋,陈友利,刘军等.基于相似方向的二叉树支持向量机多类分类算[J].四川师范大学学报:自然科学版,2008,31(6).
    [12]Platt J C,Cristianini N,Shawe-Taylor J.Large Margin DAGs for Multiclass Classification[C].nips,1999,(12).