最短特征线段多分类器系统设计

丁建坤 1; 韩德强 1; 杨艺 2

doi:10.7652/xjtuxb201509014

您当前的位置：

首页 >

文章列表页 >

最短特征线段多分类器系统设计

更新时间：2025-07-09

- 最短特征线段多分类器系统设计
- Design of Multiple Classifier Systems Based on Shortest Feature Line Segment
- 西安交通大学学报 2015年49卷第9期页码：77-83
- 作者机构：
  
  西安交通大学智能网络与网络安全教育部重点实验室,西安,710049
- 作者简介：
- 基金信息：
  
  : 2015-01-28。作者简介: 丁建坤(1990—),男,博士生;韩德强(通信作者),男,副教授,博士生导师。基金项目: 国家“973计划”资助项目(2013CB329405);国家自然科学基金资助项目(61203222);教育部博士点基金资助项目(20120201120 036);中央高校基本科研业务费专项资金资助项目(xjj2012104,xjj2014122)
- DOI：10.7652/xjtuxb201509014
  中图分类号： TP391
- 网络首发：2015-09-10，
  
  纸质出版：2015
- 稿件说明：
移动端阅览
丁建坤 1, 韩德强 1, 杨艺 2. 最短特征线段多分类器系统设计[J]. 西安交通大学学报, 2015,49(9):77-83.

Design of Multiple Classifier Systems Based on Shortest Feature Line Segment[J]. 2015, 49(9): 77-83.
丁建坤 1, 韩德强 1, 杨艺 2. 最短特征线段多分类器系统设计[J]. 西安交通大学学报, 2015,49(9):77-83. DOI： 10.7652/xjtuxb201509014.

Design of Multiple Classifier Systems Based on Shortest Feature Line Segment[J]. 2015, 49(9): 77-83. DOI： 10.7652/xjtuxb201509014.

摘要

为提高多分类器系统分类的性能

设计了一种使用最短特征线段分类器的多分类器系统。依据最短特征线段分类算法工作机理

利用特征线段长度表征样本隶属于各个类别的可能性

即模糊隶属度

对成员分类器输出形式完成由摘要级至度量级的重新建模

更多地保留输出细节以减少信息损失

进而利用基于模糊的证据融合规则实现成员分类器的度量级融合

通过隶属度到mass函数的转换

利用模糊-证据融合规则实现多分类器系统的构造

进一步提高了多分类器系统分类性能。采用人工数据集和UCI数据集设计了对比实验

实验表明

与其他邻域型分类器构造的多分类器系统相比

新多分类器系统能有效提升分类正确率。

Abstract

To improve the classification performance of multiple classifier systems

a novel multiple classifier system using shortest feature line segment(SFLS)as member classifiers is proposed. According to the SFLS's algorithmic principle on classification

the length of the shortest feature line segment is used to represent the probability

i.e.

the fuzzy membership of the query sample belonging to the corresponding class. Thus

the member classifier's output is transformed from the abstract level to the measurement level. Furthermore

by transforming the fuzzy membership function into the mass function and using fuzzy-based evidential fusion rules

the classification fusion is implemented. Compared with traditional multiple classifier systems

the proposed approach can use more detailed information for implementing more effective decision-level fusion. Experimental results show that the proposed multiple classifier systems can effectively improve classification accuracy.

关键词

Keywords

references

李晶皎, 赵丽红, 王爱侠. 模式识别 [M]. 北京: 电子工业出版社, 2010.

杨艺, 韩崇昭, 韩德强. 利用特征子空间评价与多分类器融合的高光谱图像分类 [J]. 西安交通大学学报, 2010, 44(8): 20-24.

YANG Yi, HAN Chongzhao, HAN Deqiang. Hyperspectral image classification based on feature subspace evaluation and multiple classifier fusion [J]. Journal of Xi'an Jiaotong University, 2010, 44(8): 20-24.

张成军, 阴妍, 鲍久圣, 等. 多源信息融合故障诊断方法研究进展 [J]. 河北科技大学学报, 2014, 35(3): 213-221.

ZHANG Chengjun, YIN Yan, BAO Jiusheng, et al. Research progress in fault diagnosis methods based on multi-source information fusion [J]. Journal of Hebei University of Science and Technology, 2014, 35(3): 213-221.

王风华, 韩九强, 姚向华. 一种基于虹膜和人脸的多生物特征融合方法 [J]. 西安交通大学学报, 2008, 42(2): 133-137.

WANG Fenghua, HAN Jiuqiang, YAO Xianghua. Multimodal biometric fusion approach based on iris and face [J]. Journal of Xi'an Jiaotong University, 2008, 42(2): 133-137.

张彩坡. 模糊积分及多分类器融合在医疗诊断中的应用 [D]. 天津: 天津理工大学, 2010.

BREIMAN L. Bagging predictors [J]. Machine Learning, 1996, 24(2): 123-140.

焦李成, 公茂果, 王爽, 等. 自然计算、机器学习与图像理解前沿 [M]. 西安: 西安电子科技大学出版社, 2008.

KIRA K, RENDELL L. A practical approach to feature selection [C]∥Proceedings of the 9th Machine Learning. Aberdeen, UK: Morgan Kaufmann Publishers, 1992: 249-256.

MARKO R, IGOR K. Theoretical and empirical analysis of ReliefF and RReliefF [J]. Machine Learning, 2003, 53(1/2): 23-69.

DIETTERICH T G. An experimental comparison of three methods for constructing ensembles of decision trees: bagging, boosting and randomization [J]. Machine Learning, 2000, 40(2): 139-157.

梁绍一, 韩德强, 韩崇昭. 一种基于几何关系的多分类器差异性度量及其在多分类器系统构造中的应用 [J]. 自动化学报, 2014, 40(3): 449-458.

LIANG Shaoyi, HAN Deqiang, HAN Chongzhao. A novel diversity measure based on geometric relationship and its application to design of multiple classifier systems [J]. Acta Automatica Sinica, 2014, 40(3): 449-458.

XU Lei, KRZYZAK A, SUEN C Y. Methods of combining multiple classifiers and their applications to handwriting recognition [J]. IEEE Transactions on Systems, Man and Cybernetics, 1992, 22(3): 418-435.

COVER T, HART P. Nearest neighbor pattern classification [J]. IEEE Transactions on Information Theory, 1967, 13(1): 21-27.

DU Hao, CHEN Yanqiu. Rectified nearest feature line segment for pattern classification [J]. Pattern Recognition, 2007, 40(5): 1486-1497.

HAN Deqiang, HAN Chongzhao, YANG Yi. A novel classifier based on shortest feature line segment [J]. Pattern Recognition Letters, 2011, 32(3): 485-493.

HE Yunhui. Face recognition using kernel nearest feature classifiers [C]∥2006 International Conference on Computational Intelligence and Security. Piscataway, NJ, USA: IEEE, 2006: 678-683.

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.

YANG Yi, HAN Deqiang, HAN Chongzhao. Study on feature selection based on rank-level fusion [J]. Control and Decision, 2011, 26(3): 397-401.

TACNET J M, DEZERT J. Cautious OWA and evidential reasoning for decision making under uncertainty [C]∥Proceeding of the 14th International Conference on Information Fusion. Piscataway, NJ, USA: IEEE, 2011: 1-8.

SHAFER G. A mathematical theory of evidence [M]. Princeton, USA: Princeton University Press, 1967.

HAN Deqiang, DEZERT J, TACNET J M, et al. A fuzzy-cautious OWA approach with evidential reasoning [C]∥Proceedings of the 15th International Conference on Information Fusion. Piscataway, NJ, USA: IEEE, 2012: 278-285.

VALENTE F, HERNANSKY H. Combination of acoustic classifiers based on Dempster-Shafer theory of evidence [C]∥IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway, NJ, USA: IEEE, 2007: 1129-1132.

SMETS P. Data fusion in the transferable belief model [C]∥Proceedings of the 3rd International Conference on Information Fusion. Piscataway, NJ, USA: IEEE, 2000: 21-33.

DU Hao, CHEN Yanqiu. Rectified nearest feature line segment for pattern classification [J]. Pattern Recognition, 2007, 40(5): 1486-1497.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

采用聚类特征的基本概率分配生成方法及应用

利用证据神经网络的多分类器系统构造

利用证据理论的多分类支持向量数据描述算法

融合物理信息与深度算子网络的变压器绕组热-流场快速计算方法

电子级超纯氮深冷空分流程节能优化研究