The effect of generalized discriminate analysis (GDA) to the classification of optic nerve disease from VEP signals
Küçük Resim Yok
Tarih
2008
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
PERGAMON-ELSEVIER SCIENCE LTD
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
In this paper, we have investigated the effect of generalized discriminate analysis (GDA) on classification performance of optic nerve disease from visual evoke potentials (VEPs) signals. The GDA method has been used as a pre-processing step prior to the classification process of optic nerve disease. The proposed method consists of two parts. First, GDA has been used as pre-processing to increase the distinguishing of optic nerve disease from VEP signals. Second, we have used the C4.5 decision tree classifier, Levenberg Marquart (LM) back propagation algorithm, artificial immune recognition system (AIRS), linear discriminant analysis (LDA), and support vector machine (SVM) classifiers. Without GDA, we have obtained 84.37%, 93.75%, 75%, 76.56%, and 53.125% classification accuracies using C4.5 decision tree classifier, LM back propagation algorithm, AIRS, LDA, and SVM algorithms, respectively. With GDA, 93.75%, 93.86%, 81.25%, 93.75%, and 93.75% classification accuracies have been obtained using the above algorithms, respectively. These results show that the GDA pre-processing method has produced very promising results in diagnosis of optic nerve disease from VEP signals. (C) 2007 Elsevier Ltd. All rights reserved.
Açıklama
Anahtar Kelimeler
VEP signals, Generalized discriminate analysis (GDA), C4.5 decision tree classifier, Levenberg Marquart (LM) back propagation, artificial immune recognition system, linear discriminant analysis (LDA), support vector machine (SVM), optic nerve disease
Kaynak
COMPUTERS IN BIOLOGY AND MEDICINE
WoS Q Değeri
Q2
Scopus Q Değeri
Q1
Cilt
38
Sayı
1
