| 基于后验概率SVM决策树多类的分类算法 |
| |
| 引用本文： | 王冬丽,郑建国,周彦.基于后验概率SVM决策树多类的分类算法[J].微型电脑应用,2011,27(2):40-41,48,4. |
| |
| 作者姓名： | 王冬丽 郑建国 周彦 |
| |
| 作者单位： | 1. 东华大学,上海,200051
2. 上海交通大学,上海,200240 |
| |
| 摘 要： | 后验概率支持向量机方法对孤立点和噪声具有鲁棒性,并且可以减少支持向量的数量,从而降低计算复杂度。因此,针对最近提出的快速分类算法c-BTS,引入样本的后验概率,提出了一种基于后验概率的SVM决策树算法P2BTS。实验结果证明,基于后验概率的支持向量机决策树P2BTS比c-BTS的分类精度更高,且所需的二类分类器个数减少,在一定程度上降低了P2BTS分类决策的时间和比较的次数,提高了分类效率。
|
| 关 键 词： | 支持向量机 决策树 后验概率 多类分类 |
Posterior Probability Based on Binary Tree of SVMs |
| |
| Wang Dongli,Zheng Jianguo,Zhou Yan.Posterior Probability Based on Binary Tree of SVMs[J].Microcomputer Applications,2011,27(2):40-41,48,4. |
| |
| Authors: | Wang Dongli Zheng Jianguo Zhou Yan |
| |
| Affiliation: | Wang Dongli1,Zheng Jianguo1,Zhou Yan2 (1.Glorious Sun School of Business and Management,Donghua University,Shanghai 200051,China,2.Automation Department,Shanghai Jiaotong University,Shanghai200240,China) |
| |
| Abstract: | Posterior probability based on SVM is robust against outliers and noises,even in the case of fuzzy or error class labels,while having less support vectors (SVs) and in turn decreased computational complexity.Hence,posterior probability is introduced for recently proposed fast classifier c-BTS,namely,a novel posterior probability based on binary tree of SVM (P2BTS) is given.Experimental results illustrate that P2BTS can obtain higher classification accuracy compared with c-BTS,while obviously less binary cla... |
| |
| Keywords: | Support Vector Machine Decision Tree Posterior Probability Multi-class Classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
