Please use this identifier to cite or link to this item: http://nopr.niscair.res.in/handle/123456789/42859
Title: Genetic Algorithm based Feature Selection for Classification of Focal and Non-focal Intracranial Electroencephalographic Signals
Authors: Sathish, E
Sivakumaran, N
Simon, S P
Raghavan, S
Keywords: Epilepsy;Genetic Algorithm;iEEG;k-NN;Wavelet Transform;SVM
Issue Date: Oct-2017
Publisher: NISCAIR-CSIR, India
Abstract: In classification problems, algorithm and feature selection plays a major role. The features selected may vary the classification accuracy of different classifier algorithms. So it’s necessary to apply efficient methods of feature selection before designing the classifier. In this paper, focal and non-focal intracranial Electroencephalographic (iEEG) signal classification aided by Genetic Algorithm (GA) for optimal feature selection is proposed for Support Vector Machines (SVM), k-Nearest Neighborhood (k-NN) and Feed Forward Neural Network (FFNN) classifiers. Statistical features from raw signal and Discrete Wavelet Transforms (DWT) were extracted from iEEG signal to have pool of features from which the optimal feature sets are selected by GA. The results of each classifier were compared based on their selected features and performance estimates.
Page(s): 614-619
URI: http://nopr.niscair.res.in/handle/123456789/42859
ISSN: 0975-1084 (Online); 0022-4456 (Print)
Appears in Collections:JSIR Vol.76(10) [October 2017]

Files in This Item:
File Description SizeFormat 
JSIR 76(10) 614-619.pdf296.86 kBAdobe PDFView/Open


Items in NOPR are protected by copyright, with all rights reserved, unless otherwise indicated.