Comparison between five classifiers for automatic scoring of human sleep recordings - Archive ouverte HAL Accéder directement au contenu
Chapitre D'ouvrage Année : 2005

Comparison between five classifiers for automatic scoring of human sleep recordings

Résumé

The aim of this work is to compare the performances of 5 classifiers (linear and quadratic classifiers, k nearest neighbors, Parzen kernels and neural network) to score a set of 8 biological parameters extracted from EEG and EMG, in six classes corresponding to different sleep stages. The data base is composed of 17265 epochs of 20s recorded from 4 patients. Each epoch has been classified by an expert. In order to evaluate the classifiers, learning and testing sets of fixed size are randomly drawn and are used to train and test the classifiers. After several trials, an estimation of the misclassification percentage and its variability is obtained (optimistically and pessimistically). Data transformations toward normal distribution are explored as an approach to deal with extreme values. It is shown that these transformations improve significantly the results of the classifiers based on data proximity.
Fichier non déposé

Dates et versions

hal-00123995 , version 1 (11-01-2007)

Identifiants

  • HAL Id : hal-00123995 , version 1

Citer

Guillaume Jean-Paul Claude Becq, Sylvie Charbonnier, Florian Chapotot, Alain Buguet, Lionel Bourdon, et al.. Comparison between five classifiers for automatic scoring of human sleep recordings. Saman Halgamuge and Lipo Wang. Classification and Clustering for Knowledge Discovery, Springer Verlag, pp.113, 2005. ⟨hal-00123995⟩
1572 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More