Subscription or Fee Access
Comparison of Logistic Regression, K-Nearest Neighbor, Support Vector Machine, Multilayer Perceptron, and Bagging Classification and Regression Trees Using Wavelet-Based Features in Detection of Epileptic Signals
Classification techniques are widely used to classify the various health data. One of them is electroencephalogram (EEG) signals which are the essential component for the diagnosis and analysis of epilepsy. In this study, performances of classification techniques were compared in order to classify normal and inter-ictal EEG signals which were recorded from 10 normal subjects and 10 epileptic patients. This paper uses discrete wavelet transform (DWT) as the feature extraction methods and employs logistic regression (LR), k-nearest neighbor (KNN), support vector machine (SVM), multilayer perceptron (MLP), and bagging classification and regression trees (Bagging CART) as the classifiers. Performances of classification techniques were compared using receiver operating characteristics (ROC) curve. Areas under the ROC curves are 0.984, 0.976, 0.974, 0.950, and 0.961, respectively for SVM, KNN, Bagging CART, MLP, and LR. SVM, KNN and Bagging CART were found statistically better than MLP and LR to classify normal and epileptic EEG signals in this data set according to ROC curve. This study showed that SVM, KNN and Bagging CART are capable to provide a good discrimination between normal and epileptic EEG signals.
K-Nearest Neighbor, Support Vector Machine, Multilayer Perceptron, Bagging, Classification and Regression Trees, Discrete Wavelet Transform, Epilepsy
Disclaimer/Regarding indexing issue:
We have provided the online access of all issues and papers to the indexing agencies (as given on journal web site). It’s depend on indexing agencies when, how and what manner they can index or not. Hence, we like to inform that on the basis of earlier indexing, we can’t predict the today or future indexing policy of third party (i.e. indexing agencies) as they have right to discontinue any journal at any time without prior information to the journal. So, please neither sends any question nor expects any answer from us on the behalf of third party i.e. indexing agencies.Hence, we will not issue any certificate or letter for indexing issue. Our role is just to provide the online access to them. So we do properly this and one can visit indexing agencies website to get the authentic information.