摘要： 数据挖掘对象是大型数据库中的海量数据,而数据库中记录包含众多属性,由于其中存在的冗余和不相关属性降低了数据挖掘性能,增加了算法复杂性,因此,特征子集选择问题成为数据挖掘领域中的重要研究课题。该文根据过滤法思想,提出了基于遗传算法的特征子集选择算法,实验证明该算法获得了良好的收敛性和稳定性。

关键词: 数据挖掘, LV算法, 特征子集选择, 遗传算法, 模拟退火算法

Abstract: Data mining aims at huge data in the very large database. Records in this database include many redundant and irrelated features, which degrade the data mining performance and increase the algorithm complexity. So, feature subset selection becomes one of the important research issues in the field of data mining. In this paper, an algorithm of feature subset selection based on genetic algorithm is proposed according to the filter approach. It is proved by experimental results that the convergence and the stability of this algorithm are adequately achieved.

中图分类号: 

• TP311.12