| 基于幂集演化的属性约简方法 |
| |
| 引用本文： | 潘冠宇,刘大有,刘晓华,窦全胜. 基于幂集演化的属性约简方法[J]. 吉林大学学报(工学版), 2008, 0(Z2) |
| |
| 作者姓名： | 潘冠宇 刘大有 刘晓华 窦全胜 |
| |
| 作者单位： | 吉林工商学院计算机系;吉林大学计算机科学与技术学院;山东工商学院计算机科学与技术学院; |
| |
| 基金项目： | 国家自然科学基金项目(60573073)． |
| |
| 摘 要： | 提出了一种基于幂集演化的属性约简方法,该方法作为一种基于遗传算法的属性约简的改进方法,编码方式采用在属性集的幂集空间上进行直接编码,由于其编码具有不定长的特性,因此在遗传操作策略上采取了一种与二进制编码完全不同的交叉和变异策略,使得该方法在选取合适评价函数的权重参数时,能明显加快算法的收敛速度。
|
| 关 键 词： | 人工智能 数据挖掘 粗糙集 属性约简 幂集 遗传算法 |
Reduction methods of attributes based on power set evolution |
| |
| PAN Guan-yu LIU Da-you LIU Xiao-hua DOU Quan-sheng. Reduction methods of attributes based on power set evolution[J]. Journal of Jilin University:Eng and Technol Ed, 2008, 0(Z2) |
| |
| Authors: | PAN Guan-yu LIU Da-you LIU Xiao-hua DOU Quan-sheng |
| |
| Affiliation: | 1.Department of Computer Science;Jilin Business and Technology College;Changchun 130062;China;2.College of Computer Science and Technology;Jilin University;Changchun 130012;3.School of Computer Science and Technology;Shandong Institute of Business and Technology;Yantai 264005;China |
| |
| Abstract: | An improved method based on power set evolution is proposed,which improves existing attribute reduction based on genetic algorithm.Because characteristic length is not known,we use a different approach from the crossover and mutation in genets algorithm.This method can effectively speed up the convergence rate.Our experiments show that this algorithm works well in most tested data sets. |
| |
| Keywords: | artificial intelligence data mining rough set attribute reduction power set genetic algorithm |
| 本文献已被 CNKI 等数据库收录！ |
