A one-dimensional analysis for the probability of error of linear classifiers for normally distributed classes

DOI10.1016/J.PATCOG.2004.12.002MaRDI QIDQ2485069zbMATH OpenOpenAlexFDO

Authors Luis G. Rueda

Publication date 3 August 2005

Published in Pattern Recognition (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.patcog.2004.12.002

zbMATH Keywords

Gaussian distributionsLinear discriminant analysisFisher's classifierCurse of dimensionalityError rate evaluation

Mathematics Subject Classification ID

Pattern recognition, speech recognition (68T10)

Recommendations

• scientific article; zbMATH DE number 4066167
• scientific article; zbMATH DE number 1203382
• A lower bound on the performance of the quadratic discriminant function
• On misclassification probabilities of linear and quadratic classifiers
• Toward a tight upper bound for the error probability of the binary Gaussian classification problem

Cites work

• scientific article; zbMATH DE number 47593 (Why is no real title available?)
• scientific article; zbMATH DE number 5046126 (Why is no real title available?)
• A PAC-Bayesian margin bound for linear classifiers
• A novel method for Fisher discriminant analysis.
• Algorithm 715: SPECFUN–a portable FORTRAN package of special function routines and test drivers
• An efficient approach to compute the threshold for multi-dimensional linear classifiers
• Approximations to Multivariate Normal Rectangle Probabilities Based on Conditional Expectations
• Expected classification error of the Fisher linear classifier with pseudo-inverse covariance matrix
• Numerically stable generation of correlation matrices and their factors
• The Evaluation of General Non-Centred Orthant Probabilities

Cited in

(4)

• Toward a tight upper bound for the error probability of the binary Gaussian classification problem
• A comment on: ``A one-dimensional analysis for the probability of error of linear classifiers for normally distributed classes by Rueda
• An efficient approach to compute the threshold for multi-dimensional linear classifiers
• Estimation in one–way classification with heteroscedastic error

Describes a project that uses

Uses Software

• SPECFUN
• UCI
• QSIMVN

This page was built for publication: A one-dimensional analysis for the probability of error of linear classifiers for normally distributed classes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2485069)