| 一种基于浓缩树结构的属性约简算法 |
| |
| 引用本文： | 张忠平,林志杰,李岩,陈丽萍.一种基于浓缩树结构的属性约简算法[J].计算机应用与软件,2009,26(8):45-48. |
| |
| 作者姓名： | 张忠平 林志杰 李岩 陈丽萍 |
| |
| 作者单位： | 燕山大学信息科学与工程学院,河北,秦皇岛,066004 |
| |
| 基金项目： | 教育部科学技术研究重点项目，河北省教育厅科研计划项目，国家自然科学基金项目 |
| |
| 摘 要： | 属性约简是粗糙集理论的重要研究内容之一,其中基于区分矩阵的约简算法是一种高效的约简算法,但算法具有很高的空间复杂度.为了减少区分矩阵的空间开销,利用浓缩树结构,结合区分矩阵单个属性一定为核属性的特征,提出改进的生成浓缩树算法,压缩存储区分矩阵中的非空数据项,且不丢失原区分矩阵的所有信息;利用生成的浓缩树结构结合启发式策略,给出属性约简算法.实验结果表明,算法正确有效并且空间复杂度有明显降低.
|
| 关 键 词： | 粗糙集 区分矩阵 属性约简 浓缩树 |
ALGORITHM FOR ATTRIBUTE REDUCTION BASED ON CONDENSING TREE STRUCTURE |
| |
| Zhang Zhongping,Lin Zhijie,Li Yan,Chen Liping.ALGORITHM FOR ATTRIBUTE REDUCTION BASED ON CONDENSING TREE STRUCTURE[J].Computer Applications and Software,2009,26(8):45-48. |
| |
| Authors: | Zhang Zhongping Lin Zhijie Li Yan Chen Liping |
| |
| Affiliation: | College of Information Science and Engineering;Yanshan University;Qinhuangdao 066004;Hebei;China |
| |
| Abstract: | Attributes reduction is one of the important parts studied in rough set theory.The reduction algorithms based on discernable matrix are efficient attributes reduction algorithms,but have very high space complexity.In order to reduce the storage cost of the discernable matrix,in this paper it employed the condensing tree and combined together the speciality of the discernable matrix which single attribute must be the core attribute,proposed an improved generating condensing tree algorithm to compress and sto... |
| |
| Keywords: | Rough set Discernable matrix Attribute reduction Condensing tree |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
