| 连续属性的离散化算法 |
| |
| 引用本文： | 刘文军,谷云东.连续属性的离散化算法[J].郑州大学学报(理学版),2006,38(4):41-43,55. |
| |
| 作者姓名： | 刘文军 谷云东 |
| |
| 作者单位： | 1. 长沙理工大学数计学院,长沙,410076
2. 北京师范大学管理学院,北京,100875 |
| |
| 基金项目： | 国家自然科学基金资助项目,编号70501006，长沙理工大学博士科研启动基金项目,编号52129905 |
| |
| 摘 要： | 结合聚类的思想与信息增益性质,给出一种基于距离与信息增益相结合的连续属性离散化方法.此方法不仅考虑了属性值之间的序关系,而且考虑了属性值之间的相对大小关系.此算法的一个最大优点是能自动调整离散化过程中的阈值,且能达到所要求的决策表相容度.
|
| 关 键 词： | 离散化 距离 信息增益 相容度 |
| 文章编号： | 1671-6841（2006）04-0041-03 |
| 收稿时间： | 03 15 2006 12:00AM |
| 修稿时间： | 2006-03-15 |
Discretization Algorithm of the Continuous Attribute |
| |
| LIU Wen-jun,GU Yun-dong.Discretization Algorithm of the Continuous Attribute[J].Journal of Zhengzhou University:Natural Science Edition,2006,38(4):41-43,55. |
| |
| Authors: | LIU Wen-jun GU Yun-dong |
| |
| Institution: | 1. Department of Mathematics and Computing Science, Changsha University of Science and Technology, Changsha 410076,China ; 2. Department of Management ,Beijing Normal University ,Beijing 100875 ,China |
| |
| Abstract: | Combining the philosophy of clustering and information gain,a discretization algorithm based on distance and information gain is proposed.This method thinks over not only the ordinal relation of attribute values,but also the relative distance of attribute values.The prominent merit of this algorithm is that it can adjust the threshold automatic and satisfy the consistence degree requested. |
| |
| Keywords: | discretization distance information gain consistence degree |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
