计算机与现代化

• 算法设计与分析 • 上一篇    下一篇

半动态集成选择分类方法

  

  1. 1.陕西财经职业技术学院会计二系,陕西咸阳712000;2.陕西财经职业技术学院会计一系,陕西咸阳712000
  • 收稿日期:2014-11-21 出版日期:2015-02-28 发布日期:2015-03-06
  • 作者简介:李瑞(1979),女,陕西礼泉人,陕西财经职业技术学院会计二系讲师,硕士,研究方向:软件工程; 袁小玲(1980),女,陕西财经职业技术学院会计一系讲师,硕士,研究方向:软件工程。

Semidynamic Integration Selection Classification Method

  1. 1. The Second  Accounting Department, Shaanxi Vocational College of Finance and Economics, Xianyang 712000, China; 
    2. The First Accounting Department, Shaanxi Vocational College of Finance and Economics, Xianyang 712000, China
  • Received:2014-11-21 Online:2015-02-28 Published:2015-03-06

摘要: 在集成学习领域,传统的动态集成选择需要为每一个样本选择子分类器组成集成分类器,这极大地增加了计算复杂度。针对这一问题,提出一种新的半动态集成选择方法。该方法分为两阶段,第一阶段为所有的测试样本选择最好的个体分类器组成一个集成分类器,第二阶段从剩余的个体分类器集合中为当前测试样本动态地选择子分类器组成一个集成分类器。最终的分类结果通过融合两阶段得到集成分类器的结果得到。通过对UCI数据测试的结果表明,该算法不仅能取得较好的分类性能,而且能极大地降低计算复杂度。

关键词: 集成学习, 选择性集成, 动态集成选择, 分类

Abstract: Traditional Dynamic Ensemble Selection (DES) in ensemble learning needs to select individual classifiers for all the test samples. However, it leads to highly computational cost. Due to this issue, a new Semi Dynamic Ensemble Selection (SemiDES) strategy is proposed in this paper, which consists of two stages. Individual classifiers are selected for all the test samples in the first stage. In the second stage, the classifiers for each test sample are selected dynamically. The final result is obtained by integrating the output of the two stages. The experimental results on UCI data set demonstrate the proposed method can obtain a better classification performance. Moreover, SemiDES can reduce the computational cost greatly.

Key words:  , ensemble learning; selective ensemble; dynamic ensemble selection; classification

中图分类号: