| 利用增益率构建混合型选择性不完整数据分类器 |
| |
| 引用本文： | 陈景年,黄厚宽,徐力,伊传环.利用增益率构建混合型选择性不完整数据分类器[J].北京交通大学学报(自然科学版),2009,33(5). |
| |
| 作者姓名： | 陈景年 黄厚宽 徐力 伊传环 |
| |
| 作者单位： | 山东财政学院信息与计算科学系,山东,济南,250014;北京交通大学计算机与信息技术学院,北京,100044;济南市公路管理局,山东济南,250013 |
| |
| 基金项目： | 国家自然科学基金资助项目 |
| |
| 摘 要： | 鉴于已有的绝大多数选择性分类算法主要用于完整数据,而现实中的数据通常是不完整的并且包含许多冗余属性或无关属性,本文在已有工作基础上利用信息增益率构建了一种用于不完整数据的混合型的选择性贝叶斯分类器:GBSD.在12个标准的不完整数据集上的实验结果表明,GBSD不仅能大幅度减少属性数目,而且比已有工作更能有效改善分类准确率和效率.
|
| 关 键 词： | 信息增益率 贝叶斯方法 分类 特征选择 不完整数据 |
Constructing Hybrid Selective Classifiers for Incomplete Data with Gain-Ratio |
| |
| CHEN Jingnian,HUANG Houkuan,XU Li,YIN Chuanhuan.Constructing Hybrid Selective Classifiers for Incomplete Data with Gain-Ratio[J].JOURNAL OF BEIJING JIAOTONG UNIVERSITY,2009,33(5). |
| |
| Authors: | CHEN Jingnian HUANG Houkuan XU Li YIN Chuanhuan |
| |
| Institution: | CHEN Jingnian1,HUANG Houkuan2,XU Li3,YIN Chuanhuan2(1.Dept.of Information , Computing Science,Sh,ong University of Finance,Jinan Sh,ong,250014,China,2.School of Computer , Information Technology,Beijing Jiaotong University,Beijing 100044,3.Jinan Road Management Bureau,Jinan Sh,ong 250013,China) |
| |
| Abstract: | Due to most selective classifiers mainly deal with complete data and actual data sets are often incomplete and have many redundant or irrelevant attributes,a hybrid selective classifier for incomplete data denoted as GBSD is presented in this paper.GBSD is based on former work and Information gain ratio.Experiment results on twelve standard incomplete data sets show that the GBSD not only can enormously reduce the number of attributes,but also can more effectively improve the accuracy and efficiency of clas... |
| |
| Keywords: | information gain ratio bayesian method classification feature selection incomplete data |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
