特征的支持度与其分类能力的关系研究

doi:10.3969/j.issn.0372-2112.2015.02.007

电子学报 ›› 2015, Vol. 43 ›› Issue (2): 248-254.DOI: 10.3969/j.issn.0372-2112.2015.02.007

所属专题：机器学习—特征选择

特征的支持度与其分类能力的关系研究

尹建芹^1,2, 田国会³, 魏军¹, 李金屏¹, 林佳本²

1. 济南大学信息科学与工程学院山东省网络环境智能计算技术重点实验室, 山东济南 250022;
2. 中国科学院太阳活动重点实验室, 北京 100012;
3. 山东大学控制科学与工程学院, 山东济南 250061

收稿日期:2013-11-05 修回日期:2014-04-21 出版日期:2015-02-25
- 作者简介:
- 尹建芹 女,1978年11月出生,山东潍坊人.副教授,博士.2000年获山东工业大学工学学士学位,2002年和2013年获山东大学工学硕士和工学博士学位.现工作于济南大学信息科学与工程学院,主要从事模式识别、机器学习及图像处理等相关研究. E-mail:ise_yinjq@ujn.edu.cn;田国会 男,1969年8月出生,河北河间人.教授、博士生导师.1990年、1993年和1997年分别在山东大学、山东工业大学和东北大学获理学学士、工学硕士和工学博士学位.现为山东大学服务机器人研究室主任,主要从事服务机器人、智能空间等的研究工作.
- 基金资助:
- 国家自然科学基金 (No.61075092,No.61203341,No.61173079); 山东省高等学校科技发展计划 (No.TJY1112); 中科院太阳活动重点实验室开放课题 (No.TQT1302)

Research on the Relationship of the Support and the Discriminative Ability for Classification of Features

YIN Jian-qin^1,2, TIAN Guo-hui³, WEI Jun¹, LI Jin-ping¹, LIN Jia-ben²

1. Shandong Provincial Key Laboratory of Network Based Intelligent Computing, School of Information Science and Engineering, University of Jinan, Jinan, Shandong 250022, China;
2. Key Laboratory of Solar Activity, National Astronomical Observatories, Chinese Academy of Sciences, Beijing 100012, China;
3. School of Control Science and Engineering, Shandong University, Jinan, Shandong 250061, China

Received:2013-11-05 Revised:2014-04-21 Online:2015-02-25 Published:2015-02-25
- Supported by:
- National Natural Science Foundation of China (No.61075092, No.61203341, No.61173079); Science and Technology Development Project of Higher Education Institutions in Shandong Province (No.TJY1112); Open Project of Key Laboratory of Solar Activity of Chinese Academy of Sciences (No.TQT1302)

摘要/Abstract

摘要：

频繁模式挖掘在分类问题中得到了广泛的应用,大量的工作利用频繁模式挖掘对分类问题进行特征选择,但对于为什么频繁模式挖掘可以在分类问题中进行有效的特征选择则缺乏系统的研究.为了为频繁模式挖掘在分类问题中的特征选择应用提供理论基础,需要确立特征的支持度与特征分类能力之间的关系,本文以特征的信息增益作为分类能力的评价准则,讨论其与特征支持度之间的联系.首先证明了信息增益是特征支持度的上凸函数;然后,在二类问题和多类问题情况下,分别证明了具有低支持度或高支持度的特征具有有限的信息增益,即具有低支持度或高支持度的特征具有有限的分类能力.最后,通过仿真实验验证了支持度与信息增益之间的关系,为频繁模式挖掘在分类问题中的应用提供了理论基础.

关键词: 频繁模式, 分类, 特征选择, 信息增益

Abstract:

Frequent pattern mining is used widely in feature selection for classification problem.In order to provide theoretical basis for the application,we established the relationship between the classification discriminative ability and the support of the feature.Information gain was adopted as evaluation criteria,and we discussed the connection between the support of the feature and its discriminative ability.Firstly,we proved the information gain is a concave function about the support of the feature;secondly,we proved the conclusion that the feature with too-high or too-low support has limited discriminative ability under the two classes and multiple classes circumstances separately;Finally,simulation experiments validate our conclusions.And the conclusion provides a theoretical basis for the application of frequent pattern mining in classification problems.

Key words: frequent pattern, classification, feature selection, information gain

中图分类号:

TP181

尹建芹, 田国会, 魏军, 等. 特征的支持度与其分类能力的关系研究[J]. 电子学报, 2015, 43(2): 248-254.

YIN Jian-qin, TIAN Guo-hui, WEI Jun, et al. Research on the Relationship of the Support and the Discriminative Ability for Classification of Features[J]. Acta Electronica Sinica, 2015, 43(2): 248-254.

参考文献

[1] 陈晓云,陈,王雷,李荣陆,胡运发.基于分类规则树的频繁模式文本分类[J].软件学报,2006,17 (5):1017-1025. Chen Xiaoyun,Chen Yi,Wang Lei,Li Ronglu,Hu Yunfa.Text categorization based on classification rules tree by frequent patterns[J].Journal of Software,2006,17(5):1017-1025.(in Chinese)
[2] H Lodhi,C Saunders,J Shawe-Taylor,N Cristianini,C Watkins.Text classification using string kernels[J].Journal of Machine Learning Research,2002,2(3):419-444.
[3] Y Li,S M Chung,J D Holt.Text document clustering based on frequent word meaning sequences[J].Data and Knowledge Engineering,2008,64(1):381-404.
[4] 赵建邦,董安国,高琳.一种用于生物网络数据的频繁模式挖掘算法[J].电子学报,2010,38(8):1803-1807. Zhao Jianbang,Dong Anguo,Gao Lin.An algorithm for frequent pattern mining in biological networks[J].Acta Electronica Sinica,2010,38(8):1803-1807.(in Chinese)
[5] Young-Rae Cho,Aidong Zhang.Predicting protein function by frequent functional association pattern mining in protein interaction networks[J].IEEE Transactions on Information Technology in Biomedicine,2010,14(1):30-36.
[6] R Alves,D R Baena,J S A Ruiz.Gene association analysis:a survey of frequent pattern mining from gene expression data[J].Briefings in Bioinformatics,2010,11(2):210-224.
[7] Hong Cheng,Xifeng Yan,Jiawei Han,Chih-Wei Hsu.Discriminative frequent pattern analysis for effective classification[A].In:IEEE 23rd International Conference on Data Engineering[C].Istanbul,Turkey,2007,716-725.
[8] Han Jiawei,Cheng Hong,Xin Dong,Yan Xifeng.Frequent pattern mining:current status and future directions[J].Journal of Data Mining and Knowledge Discovery,2007,15(1):55-86.
[9] H Carl,F John.Sequential pattern mining-approaches and algorithms[J].ACM Computing Surveys,2013,45(2):1-19.
[10] 高琳,覃桂敏,周晓峰.图数据中频繁模式挖掘算法研究综述[J].电子学报,2008,36(8):1603-1609. Lin Gao,Guimin Qin,Xiaofeng Zhou.An overview of algorithms for mining frequent patterns in graph data[J].Acta Electronica Sinica,2008,36(8):1603-1809.(in Chinese)
[11] 万里,廖建新,朱晓民,倪萍.一种基于频繁模式的时间序列分类框架[J].电子与信息学报,2010,32(2):261-266. Li Wan,Jianxin Liao,Xiaomin Zhu,Ping Ni.A frequent pattern based times series classification framework[J].Journal of Electronics and Information Technology,2010,32(2):261-266.(in Chinese)
[12] Lee Jae-Gil,Han Jiawei,Li Xiaolei,Cheng Hong.Mining discriminative patterns for classifying trajectories on road networks[J].IEEE Transactions on Knowledge and Data Engineering,2011,23(5):713-726.
[13] C M Bishop.Pattern Recognition and Machine Learning[M].Springer Press,2006:55-57.
[14] B Stephen,L Vandenberghe.Convex Optimization[M].Cambridge University Press,England.2004:136-146.
[15] R S Michalski,I Mozetic,J Hong,N Lavrac.The multi-purpose incremental learning system AQ15 and its testing application to three medical domains[A].In Proceedings of the Fifth National Conference on Artificial Intelligence[C].Philadelphia,America.1986.1041-1045.
[16] I Guyon,R Steve Gunn,A Ben-Hur,G Dror.Result analysis of the NIPS 2003 feature selection challenge[A].Proceedings on Advances in Neural Information Processing Systems[C].Vancouver,Canada.2004.545-552.
[17] A R Rocha,R Sousa,G A Barreto,J S Cardoso.Diagnostic of pathology on the vertebral column with embedded reject option[A].Proceedings of the 5th Iberian Conference on Pattern Recognition and Image Analysis[C].Grancanaria,Spain.2011.588-595,

特征的支持度与其分类能力的关系研究

Research on the Relationship of the Support and the Discriminative Ability for Classification of Features

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	吕杭, 蒋明峰, 李杨, 张鞠成, 王志康. 基于混合时频域特征的卷积神经网络心律失常分类方法的研究[J]. 电子学报, 2023, 51(3): 701-711.
[2]	王子为, 鲁继文, 周杰. 基于自适应梯度优化的二值神经网络[J]. 电子学报, 2023, 51(2): 257-266.
[3]	许新征, 李杉. 基于特征膨胀卷积模块的轻量化技术研究[J]. 电子学报, 2023, 51(2): 355-364.
[4]	李雪莹, 王田路, 梁鹏, 王翀. 基于系统模型的用户评论中非功能需求的自动分类[J]. 电子学报, 2022, 50(9): 2079-2089.
[5]	杨小龙, 李欣玥, 周牧, 王勇, 何维. 基于多维模糊映射AP优化的WLAN室内定位方法[J]. 电子学报, 2022, 50(8): 1875-1884.
[6]	江兵兵, 何文达, 吴兴宇, 项俊浩, 洪立斌, 盛伟国. 基于自适应图学习的半监督特征选择[J]. 电子学报, 2022, 50(7): 1643-1652.
[7]	陆克中, 陈超凡, 蔡桓, 吴定明. 面向概念漂移和类不平衡数据流的在线分类算法[J]. 电子学报, 2022, 50(3): 585-597.
[8]	王雪松, 张翰林, 程玉虎. 基于自编码器和超图的半监督宽度学习系统[J]. 电子学报, 2022, 50(3): 533-539.
[9]	肖进胜, 张舒豪, 陈云华, 王元方, 杨力衡. 双向特征融合与特征选择的遥感影像目标检测[J]. 电子学报, 2022, 50(2): 267-272.
[10]	赵旭东, 陶然, 李伟, 张蒙蒙. 高光谱图像分数域信息提取理论与方法进展[J]. 电子学报, 2022, 50(12): 2874-2883.
[11]	孙晨峰, 吕卫民, 戴洪德, 张浩晨. 一种基于TimeGAN和OCSVM的多元退化设备小子样数据增广方法[J]. 电子学报, 2022, 50(11): 2678-2687.
[12]	刘浩阳, 林耀进, 刘景华, 吴镒潾, 毛煜, 李绍滋. 由粗到细的分层特征选择[J]. 电子学报, 2022, 50(11): 2778-2789.
[13]	高雷阜, 张梦瑶, 赵世杰. 融合簇边界移动与自适应合成的混合采样算法[J]. 电子学报, 2022, 50(10): 2517-2529.
[14]	管永明, 王刚, 骆凯波, 吕梁, 吕晓雯, 史玉良. 基于极端不平衡学习的泛化低压异常箱表关系识别研究与应用[J]. 电子学报, 2021, 49(8): 1507-1514.
[15]	杨国为, 万鸣华, 赖志辉, 张凡龙, 黄伟强. 具有合适拒识机制的高正确识别率分类器设计[J]. 电子学报, 2021, 49(8): 1569-1576.