期刊文献+

MFA-DMFS:一种新的多分类器融合方法及其应用研究 被引量:1

MFA-DMFS:new multi-classifier fusion method and its application
在线阅读 下载PDF
导出
摘要 针对分类器的构建,在保证基分类器准确率和差异度的基础上,提出了采用差异性度量特征选择的多分类器融合算法(multi-classifier fusion algorithm based on diversity measure for feature selection,MFA-DMFS)。该算法的基本思想是在原始特征集中采用Relief特征评估结果按权值大小选择特征,构造特征子集,通过精调使各特征子集间满足一定的差异性,从而构建最优的基分类器。MFA-DMFS不但能提高基分类器的准确率,而且保持基分类器间的差异,克服差异性和平均准确率之间存在的相互制约,并实现这两方面的平衡。在UCI数据集上与基于Bagging、Boosting算法的多分类器融合系统进行了对比实验,实验结果表明,该算法在准确率和运行速度方面优于Bagging和Boosting算法,此外在图像数据集上的检索实验也取得了较好的分类效果。 The fusion of multiple classifiers is an important means of improving the efficiency of pattern recognition.The text,from the viewpoint of the accuracy of the component classifier as well as weight differences between the two classifiers,proposed that the multi-classifier fusion algorithm,based on differences in measurement,fell into a two-stage classifiers used to build the characteristics of the different components of a subset,the relief characteristics of the assessment results in accordance with the weights,were firstly used,from the original feature set to choose features to make into feature subset selection,and then by the fine-tuning to enable the feature subset to meet certain differences.Compared with the result of the experiments,it was found that the accuracy of the algorithm is always better than Bagging and the multiple classifiers fusion system constructed by Boosting algorithm.Moreover,its running speed is much higher than Bagging and Boosting algorithms.Finally,it was found that,in image database retrieval experiments,it has achieved better classification results.
出处 《计算机应用研究》 CSCD 北大核心 2012年第2期522-526,共5页 Application Research of Computers
基金 国家自然科学基金资助项目(51108209 50875112 70972048) 江苏省自然科学基金资助项目(BK2010339) 江苏省高校自然科学基金资助项目(10KJD580001) 江苏省博士创新基金资助项目(CXLX11_0593) 镇江市社会发展项目(FZ2008046)
关键词 特征选择 差异性度量 分类器融合 图像检索 PCA 仿真 feature selection difference metric classifier fusion image retrieval principle component analysis simulation
作者简介 作者简介:梁军(1976-),男(回族),江苏扬州人,副教授,博士研究生,主要研究方向为智能交通、模式识别、多agent理论与应用(liangjun@ujs.edu.cn); 陈龙(1958-),男,江苏靖江人,教授,博导,主要研究方向为车辆动态设计模拟及控制; 汪若尘(1977-),男,河南信阳人,副教授,主要研究方向为车辆动态设计模拟及控制; 胥正川(1973-),男,江苏镇江人,副教授,博士(后),主要研究方向为移动商务; 胥杜杰(1975-),男,四川成都人,助理工程师,主要研究方向为信息系统与信息管理.
  • 相关文献

参考文献13

  • 1孙亮,韩崇昭,沈建京,戴宁.集成特征选择的广义粗集方法与多分类器融合[J].自动化学报,2008,34(3):298-304. 被引量:10
  • 2刘汝杰,袁保宗,唐晓芳.一种新的基于聚类的多分类器融合算法[J].计算机研究与发展,2001,38(10):1236-1241. 被引量:12
  • 3杨庚,王安琪,陈正宇,许建,王海勇.一种低耗能的数据融合隐私保护算法[J].计算机学报,2011,34(5):792-800. 被引量:58
  • 4杨艺,韩德强,韩崇昭.基于排序融合的特征选择[J].控制与决策,2011,26(3):397-401. 被引量:13
  • 5XU L, KRZYZAK A, SUEN C Y. Methods of combining muhir, le classifiers and their applications to handwriting recognition[ J1. IEEE Trans on Systems, Man, and Cybernetics, 1992,22 ( 3 ) :418- 435.
  • 6彭凯,秦永彬,许道云.基于逻辑回归的客户稳定度建模[J].计算机工程,2011,37(9):12-15. 被引量:7
  • 7HOT K, HULl, J J, SRIHARI S N. Decision combination in muhiple classifier systems[J]. IEEE Trans on Pattern Analysis and Ma- chino Intelligence, 1994,16(6) :66-75.
  • 8KI'ITLER J. Improving recognition rates by classifier combination: a theoretical framework [ J ]. Progress in Handwriting Recognition World Scientific, 1997,35 ( 3 ) :231 - 248.
  • 9DIETI'ERICH T. An experimental comparison of three methods for construction ensembles of decision trees: bagging, boosting, and ran- domization [ J ]. Machine Learning, 2000,40 ( 2 ) : 139-157.
  • 10BROWN G, WYATI" J, HARRIS R, et al. Diversity creation meth- otis: a survey and categorization [ J ]. Information Fusion: A spe- cial issue on diversity in multiple classifier system,2005,6 ( 1 ) : 5-20.

二级参考文献48

  • 1胡健萍.电信企业客户忠诚度模型的构建[J].科技经济市场,2008(7):23-24. 被引量:2
  • 2Sun Liang,Han Chongzhao.Dynamic weighted voting for multiple classifier fusion:a generalized rough set method[J].Journal of Systems Engineering and Electronics,2006,17(3):487-494. 被引量:9
  • 3钱锋,徐麟文.基于数据挖掘的客户忠诚度提升[J].商场现代化,2006(09S):46-47. 被引量:4
  • 4肖迪,胡寿松.实域粗糙集理论及属性约简[J].自动化学报,2007,33(3):253-258. 被引量:32
  • 5TAN K, OAKLEY J P.Enhancement of color image in poor visibility condition[C]//Proc of IEEE International Conference on Image Processing.2000:788-791.
  • 6NARASIMHAN S G, NAYAR S K.Contrast restoration of weather degraded images[J].IEEE Trans on Pattern Analysis and Machine Intelligence,2003,25(6):715-720.
  • 7NARASIMHAN S G, NAYAR S K.Interactive (de)weathering of an image using physical model[C]//Proc of the ICCV Workshop on Color and Photometric Methods in Computer Vision.2003:3-7.
  • 8JOBSON D J, RAHMAN Z U, WOODELL G A.Properties and performance of a center/surround Retinex[J].IEEE Trans on Image Processing,1997,6(3):456-462.
  • 9JOBSON D J, RAHMAN Z U,WOODELL G A.A multiscale Retinex for bridging the gap between color images and the human observation of scenes[J].IEEE Trans on Image Processing,1997,6(7):969-976.
  • 10RAHMAN Z U, JOBSON D J , WOODELL G A.Retinex processing for automatic image enhancement[J].Journal of Electronic Imaging,2004,13(1):100-107.

共引文献99

同被引文献8

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部