Comparative analysis of statistical pattern recognition methods in high dimensional settings

An extensive simulation study is reported comparing eight statistical classification methods, focusing on problems where the number of observations is less than the number of variables. Using a wide range of artificial and real data sets, two types of classifiers are contrasted; methods that classif...

Full description

Saved in:
Bibliographic Details
Published inPattern recognition Vol. 27; no. 8; pp. 1065 - 1077
Main Authors Aeberhard, Stefan, Coomans, Danny, de Vel, Olivier
Format Journal Article
LanguageEnglish
Published Oxford Elsevier Ltd 01.08.1994
Elsevier Science
Subjects
Applied sciences
Artificial intelligence
Classifier evaluation
Computer science; control theory; systems
Dimensionality
Discriminant analysis
Exact sciences and technology
High dimensionality
Pattern recognition. Digital image processing. Computational geometry
Reduction
Simulation
Discriminant analysis
Dimensionality
Reduction
High dimensionality
Simulation
Classifier evaluation
Evaluation
Graph
Classification
Classifier
Pattern recognition
Online AccessGet full text

Cover

More Information
Summary:An extensive simulation study is reported comparing eight statistical classification methods, focusing on problems where the number of observations is less than the number of variables. Using a wide range of artificial and real data sets, two types of classifiers are contrasted; methods that classify using all variables, and methods that first reduce the number of dimensions to two or three. The simulations identified regularized discriminant analysis as the overall clearly most powerful classifier, and show that in most cases, a reduction of the dimensionality to two or three dimensions prior to classification increases the error in allocating test observations.
ISSN:0031-3203
1873-5142
DOI:10.1016/0031-3203(94)90145-7