摘要
指出现有粗糙集属性约简算法的不足,考虑并行遗传算法在处理大型数据库上的特有优势,将粗糙熵作为粗糙集不确定性的度量,给出一种求解信息系统约简集的三群体并行遗传算法。最后通过实例计算表明该算法能快速有效求解属性约简,而且对大规模数据样本的信息系统效果更为明显。
Reduction of attribute is one of the important topics in the search of rough set theory. Although many algo-rithms for reduction of attribute have been proposed, most of them have some defects. On the other hand, parallel genetic algorithm has some advantages to deal with huge data sets. In this paper, rough entropy is used to measure the uncertainties of rough set. Then a new three population parallel genetic algorithm is presented to solve reduction of attribute from data sets. It is testified by the experiment that the algorithm is effective and it is faster than ordinary algo-rithms, especially for huge data sets.
出处
《计算机科学》
CSCD
北大核心
2008年第3期219-221,共3页
Computer Science
基金
广西大学科研基金项目(No.X032016)资助
关键词
粗糙集
粗糙熵
属性约简
并行遗传算法
Rough set, Rough entropy, Reduction of attribute, Parallel genetic algorithm
