| 基于广义差别矩阵的核和属性约简算法 |
| |
| 引用本文： | 杨明,杨萍.基于广义差别矩阵的核和属性约简算法[J].控制与决策,2008,23(9). |
| |
| 作者姓名： | 杨明 杨萍 |
| |
| 作者单位： | 南京师范大学数学与计算机科学学院,南京,210097 |
| |
| 基金项目： | 国家自然科学基金，江苏省自然科学基金，江苏省高校自然科学基金 |
| |
| 摘 要： | 属性约简是粗糙集理论的重要研究内容.为此引入广义差别矩阵,提出基于广义差别矩阵的核和属性约简算法.该框架可有效避免连续属性值离散化,且有利于与其他机器学习方法相结合.理论分析表明,所提出的算法是有效而可行的.
|
| 关 键 词： | 租糙集 广义差别矩阵 核 属性约简 |
Algorithms based on general discernibility matrix for computation of a core and attribute reduction |
| |
| YANG Ming,YANG Ping.Algorithms based on general discernibility matrix for computation of a core and attribute reduction[J].Control and Decision,2008,23(9). |
| |
| Authors: | YANG Ming YANG Ping |
| |
| Affiliation: | YANG Ming,YANG Ping(School of Mathematics , Computer Science,Nanjing Normal University,Nanjing 210097,China.) |
| |
| Abstract: | Attributes reduction is one of important parts researched in rough set theory.Therefore,in this paper,after proposing a concept of general discernibility matrix,both computation of a core and attribute reduction algorithms based on general discernibility matrix are introduced.The newly proposed framework can effectively avoid discretizing the continuous-valued attributes and be easily incorporated into other machine learning methods.Theoretical analysis shows the effectiveness of the algorithm. |
| |
| Keywords: | Rough set General discernibility matrix Core Attributes reduction |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
