• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2013, Vol. 35 ›› Issue (8): 168-173.

• 论文 • 上一篇    下一篇

基于特征选择技术的集成方法研究

曹彦,王倩,周驰   

  1. (1.周口师范学院计算机科学与技术学院,河南 周口 466001;2.许昌供电公司,河南 许昌 461000)
  • 收稿日期:2012-06-01 修回日期:2012-10-08 出版日期:2013-08-25 发布日期:2013-08-25
  • 基金资助:

    中国青年基金重点项目(2012QNA01)

Research on feature selection and its ensemble method          

CAO Yan,WANG Qian,ZHOU Chi   

  1. (1.School of Computer Science and Technology,Zhoukou Normal University,Zhoukou 466001;
    2.Xuchang Power Electrical Supply Company,Xuchang  461000,China)
  • Received:2012-06-01 Revised:2012-10-08 Online:2013-08-25 Published:2013-08-25

摘要:

随着计算机与网络技术的快速发展,大数据集的出现致使人们获取的信息量正在以前所未有的速度日益剧增,如何快速获取有用信息倍受人们关注。针对如何有效剔除冗余数据问题,运用具有良好泛化能力的支持向量机的特征选择和集成分类器新技术,在支持向量机分类的基础上,以特征选择和基于特征选择的集成学习方法为主要研究内容,以具有较高分类效果的RGS算法为基础,对多个成员分类器的集成进行深入研究,并提出了RGSE算法。最后,用实验表明了算法的正确性和有效性。

关键词: 特征选择, 集成方法, 支持向量机, 遗传算法, ReliefF算法

Abstract:

With the rapid development of computer and network technology, the emergences of large data sets make the amount of information people obtain increases at an unprecedented speed. How to obtain useful information quickly are becoming people’s concerns. To solve the problem, we study on feature selection and ensemble classifiers based on support vector machine which has good generalization ability. Using RGS algorithm that has higher classification results and the technique of ensemble classifiers, RGSE algorithm is proposed. Finally, experiments demonstrate the correctness of the algorithm.

Key words: feature selection;ensemble classifiers;support vector machine;genetic algorithm;ReliefF algorithm