期刊文献+

支持向量机中核函数的性能评价策略 被引量:4

Performance Evaluation Strategy of Kernel Function for Support Vector Machine
下载PDF
导出
摘要 继神经网络方法之后,支持向量机成为机器学习领域中的有效方法,但是核函数的评价和选取问题一直存在。该文从结构风险出发,通过经验风险和置信区间2个方面对核函数的性能进行量化,给出评价核函数性能的公式,指出传统经验风险定义的缺陷,并提出了一个新的定义。实验证明了该算法的可行性和有效性。  Support vector machine algorithm becomes another important technique after neural networks in the field of machine learning,but the evaluation and choice for kernel function is not solved.Based on the structural risk theory,a quantity estimation is proposed though empirical risk and confidence interval,and an evaluation formula for kernel function is given.This article points out the default of traditional definition of empirical risk,and gives a new definition.The results of simulation experiment show the feasibility and effectiveness of the method.
出处 《计算机工程》 CAS CSCD 北大核心 2007年第19期186-187,198,共3页 Computer Engineering
基金 上海市特种光纤重点实验室科研项目(20050926) 四川省教育厅青年基金资助项目(2006B119)
关键词 核函数 支持向量机 线性可分度 线性密集度 结构风险 kernel function support vector machine linearly separable degree linearly dispersion degree structural risk
  • 相关文献

参考文献4

  • 1Vladimir N V.统计学习理论[M].许建华,张学工,译.北京:电子工业出版社,2004.
  • 2张铃.基于核函数的SVM机与三层前向神经网络的关系[J].计算机学报,2002,25(7):696-700. 被引量:54
  • 3Fisher R A. The Use of Multiple Measurements in Taxonomic Problems[J]. Annals of Eugenics, 1936, 7(2): 179-188.
  • 4Joachims T, Making Large-scale SVM Learning Practical[M]. Cambridge, UA: MIT Press, 1999,169-184.

二级参考文献1

共引文献57

同被引文献29

  • 1Wan V,Rends S.Speaker Verification Using Sequence Discriminant Support Vector Machines[J].Speech and Audio Processing,2005,13(2):203-210.
  • 2Campbell W M,Sturim D E,Reynolds D A,et al.SVM Based Speaker Verification Using a GMM Super Vector Kernel and NAP Variability Compensation[C]//Proc.of ICASSP'06.Toulouse,France:IEEE Press,2006:97-100.
  • 3Vogt R,Sridharan S.Explicit Modeling of Session Variability for Speaker Verification[J].Computer Speech and Language,2008,22(1):17-38.
  • 4Vapnik V N.统计学习论理的本质[M].张学工,译.北京:清华大学出版社,2000.
  • 5CristianiniN Shawe-TaylorJ 李国正译.支持向量机导论[M].北京:电子工业出版社,2004..
  • 6Kong Aik Lee,Chang Huai You,Haizhou Li,et al.Using DiscreteProbabilities with Bhattacharyya Measure for SVM-Based Speaker Veri-fication[J].IEEE Transactions on Audio,Speech,and LanguageProcessing,2011,19(4):861-869.
  • 7Yuxiang Shan,Jia Liu.Robust speaker recognition in cross-channelcondition based on Gaussian mixture model[J].Multimedia tools andapplications,2011,52(1):159-173.
  • 8Dalei Wu,Ji Li,Haiqing Wu.α-gaussian Mixture Modelling ForSpeaker Recognition[J].Pattern Recognition Letters,2009,30(6):589-594.
  • 9Fujimoto M,Riki Y A.Robust speech recognition in additive andchannel noise environments using GMM and EM algorithm[C].Acous-tics,Speech,and Signal Processing 2004:I-941-944.
  • 10Vijendra Raj Apsingekar,Phillip L,De Leon.Speaker Model Cluste-ring for Efficient Speaker Identification in Large Population Applica-tions[J].IEEE Transactions on Audio,Speech,and Language pro-cessing,2009,17(4):848-853.

引证文献4

二级引证文献24

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