Epileptic EEG signals recognition based on wavelet package and least square support vector machine

In order to extract the feature of epileptic EEG efficiently, and to improve the classification accuracy, a nonlinear feature extraction method based on wavelet packet Transform(WPT) and support vector machine(SVM) is proposed. The Samples are composed of five hundred EEG Public datum which include the Period of epileptic seizures. Character vectors which reflect different state of EEG signals are extracted from different frequency segments with the technology of wavelet packet decomposition which have the trait of arbitrary distinction and decomposition. The classifier is composed of the least square SVM(LSSVM) which trained by the characteristic vectors,its parameters are optimized by genetic algorithm(GA) and particle swarm optimization(PSO). Experimental results demonstrate that the classifier has good classification and generalization abilities, the identification rate of SVM which parameters are optimized by PSO algorithm reaches 91.50%.