Study on a GA-based SVM Decision-tree Multi-Classification Strategy

LIAN Ke; HUANG Jian-guo; WANG Hou-jun; LONG Bing

您当前的位置：

首页 >

文章列表页 >

Study on a GA-based SVM Decision-tree Multi-Classification Strategy

更新时间：2025-07-16

- Study on a GA-based SVM Decision-tree Multi-Classification Strategy
- Acta Electronica Sinica Vol. 36, Issue 8, Pages: 1502-1507(2008)
- 作者机构：
  
  电子科技大学自动化工程学院,四川,成都,610054
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP39
- Published：2008
- 稿件说明：
移动端阅览
LIAN Ke, HUANG Jian-guo, WANG Hou-jun, et al. Study on a GA-based SVM Decision-tree Multi-Classification Strategy[J]. Acta Electronica Sinica, 2008, 36(8): 1502-1507.
DOI：

LIAN Ke, HUANG Jian-guo, WANG Hou-jun, et al. Study on a GA-based SVM Decision-tree Multi-Classification Strategy[J]. Acta Electronica Sinica, 2008, 36(8): 1502-1507. DOI：

摘要

提出了一种基于遗传算法(GA)的SVM最优决策树生成算法

并将其应用于解决SVM多分类问题.首先以最大分类间隔为准则

利用遗传算法对传统的SVM决策树进行优化

生成最优(或近优)决策二叉树;然后在各个决策节点

利用传统的SVM算法进行二值分类

最终实现SVM的多值分类.理论分析及实验结果表明

新方法比传统的DT-SVM、DAG-SVM方法有更高的分类精度

比经典的1-a-1、1-a-r有更高的训练和分类效率.

Abstract

This paper presents a GA-based decision-tree algorithm to deal with SVM multi-class classification problem.First

GA is used to create optimal or near-optimal decision-tree automatically

which makes the margin between two classes maximal at every decision node.Then at every decision node

standard SVM is used to make binary classification.Finally

the SVM decision tree achieves multi-classification.Theoretical analysis and experiments show that the proposed method is more precise than the traditional "DT-SVM" and "DAG-SVM" methods

and has higher training and testing efficient than classical "1-a-1" and "1-a-r" methods.

关键词

Keywords

references

Views

2283

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Study of Tumor Molecular Prediction Model Based on Gene Expression Profiles

Test Case Generation and Reuse Based on Support Vector Machine Regression Model

Structure-Wise Feature Reconstruction for Hyperspectral Image Classification

Genetic Algorithm Based MIMO Equalization Parameter Optimization Technology for Mode-Division Multiplexed System

An Absorbing Structure Covering the Whole UHF Band

Related Author

LI Hui

WANG Jin-lian

QIAN Zhong-sheng

YU Qing-yuan

SONG Tao

ZHU Yi-min

ZHU Jie

ZHAO Chang

Related Institution

College of Computer Science and Technology, Beijing University of Technology

College of Electronic Information and Control Engineering,Beijing University of Technology

College of Computer Science and TechnologyBeijing University of TechnologyBeijing 100124China

College of Electronic Information and Control EngineeringBeijing University of TechnologyBeijing 100124China

School of Information Management， Jiangxi University of Finance & Economics

⁰