STRUCTURE SELECTION FOR DAG-SVM BASED ON MISCLASSIFICATION COST MINIMIZATION

Hui Yi; Xiaofeng Song; Bin Jiang

首页> 外文期刊>International Journal of Innovative Computing Information and Control >STRUCTURE SELECTION FOR DAG-SVM BASED ON MISCLASSIFICATION COST MINIMIZATION

【24h】

STRUCTURE SELECTION FOR DAG-SVM BASED ON MISCLASSIFICATION COST MINIMIZATION

机译：基于错误分类成本最小化的DAG-SVM结构选择

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相似文献
相关主题

摘要

DAG-SVM is one of the most popular multi-class support vector machines, but it has k! kinds of decision structures while dealing with k-class problems. How to select an appropriate structure is a key problem in DAG-SVM. In this paper, the classification bias of DAG-SVM was first discussed, and then the concept of misclassification cost was introduced in structure selection for DAG-SVM. Based on the principle of Misclassification Cost Minimization, a structure selection method for DAG-SVM was proposed to minimize the risk of total misclassification cost. The proposed method was tested on several benchmark data sets, and the results indicated that MCM-based selecting method for DAG-SVM obtained satisfactory results.

机译：DAG-SVM是最流行的多类支持向量机之一，但它有k！各种决策结构，同时处理k类问题。如何选择合适的结构是DAG-SVM中的关键问题。本文首先讨论了DAG-SVM的分类偏差，然后在DAG-SVM的结构选择中引入了误分类成本的概念。基于错误分类成本最小化的原理，提出了一种DAG-SVM的结构选择方法，以最大程度地降低错误分类总成本的风险。在几种基准数据集上对该方法进行了测试，结果表明基于MCM的DAG-SVM选择方法取得了满意的结果。

著录项

来源
《International Journal of Innovative Computing Information and Control》 |2011年第9期|p.5133-5143|共11页
作者
Hui Yi; Xiaofeng Song; Bin Jiang;
展开▼
作者单位

College of Automation Engineering Nanjing University of Aeronautics and Astronautics No. 29, Yudao Street, Baixia Distinct, Nanjing 210016, P. R. China;

College of Automation Engineering Nanjing University of Aeronautics and Astronautics No. 29, Yudao Street, Baixia Distinct, Nanjing 210016, P. R. China;

College of Automation Engineering Nanjing University of Aeronautics and Astronautics No. 29, Yudao Street, Baixia Distinct, Nanjing 210016, P. R. China;

展开▼
收录信息
原文格式 PDF
正文语种 eng
中图分类
关键词
misclassification cost; DAG-SVM; structure selection;

机译：分类费用;DAG-SVM;结构选择;

相似文献

外文文献
中文文献
专利

1. Cost-sensitive business failure prediction when misclassification costs are uncertain: A heterogeneous ensemble selection approach [J] . European Journal of Operational Research . 2020,第2期

机译：错误分类成本不确定时的成本敏感的业务故障预测：异构集合选择方法
2. A fast linear-in-the-parameters classifier construction algorithm using orthogonal forward selection to minimize leave-one-out misclassification rate [J] . X. HONG, S. CHEN, C. J. HARRIS International journal of systems science . 2008,第2期

机译：使用正交正向选择的快速线性参数分类器构造算法，以最大程度地减少遗漏式错误分类率
3. A Normalized Probabilistic Expectation-Maximization Neural Network for Minimizing Bayesian Misclassification Cost Risk [J] . Parag C. Pendharkar Neural processing letters . 2013,第3期

机译：最小化贝叶斯错误分类成本风险的归一化概率期望最大化神经网络
4. CMO-SMOTE: Misclassification Cost Minimization Oriented Synthetic Minority Oversampling Technique for Imbalanced Learning [C] . Changsheng Zhou, Bin Liu, Shihai Wang International Conference on Intelligent Human-Machine Systems and Cybernetics . 2016

机译：CMO-SMOTE：针对不平衡学习的错误分类成本最小化综合少数族裔过采样技术
5. Decision Support System (DSS) for machine selection: A cost minimization model [D] . Mendez Pinero, Mayra I. 2009

机译：用于机器选择的决策支持系统（DSS）：成本最小化模型
6. The use of misclassification costs to learn rule-based decision support models for cost-effective hospital admission strategies. [O] . R. Ambrosino, B. G. Buchanan, G. F. Cooper, 1995

机译：使用分类错误的成本来学习基于规则的决策支持模型以制定具有成本效益的医院入院策略。
7. A fast linear-in-the-parameters classifier construction algorithm using orthogonal forward selection to minimize leave-one-out misclassification rate [O] . Hong X., Chen S., Harris C.J. 2008

机译：一种快速线性参数分类器构造算法，使用正交前向选择来最小化一次性错误分类率
8. ON MINIMIZING THE PROBABILITY OF MISCLASSIFICATION FOR LINEAR FEATURE SELECTION [R] . L. F. Guseman, Homer F. Walker 1973

机译：最小化线性特征选择的误分类概率

STRUCTURE SELECTION FOR DAG-SVM BASED ON MISCLASSIFICATION COST MINIMIZATION

摘要

著录项

相似文献

相关主题

期刊订阅