Genetic algorithm model selection in a Fuzzy support vector machine for Automated Seizure Detection

Fuzzy Support vector machines (FSVM) have in recent years been gainfully used in various pattern recognition applications. As with any classification technique, appropriate choice of the kernels and input features play an important role in FSVM performance. In this study, an evolutionary scheme searches for optimal kernel types and parameters for automated seizure detection. We consider the Lyapunov exponent, fractal dimension and wavelet entropy for possible feature extraction. The classification accuracy of this approach is examined by applying the MIT1 Dataset and comparing results with the ANFIS and SVM. The MIT-BIH dataset has the electrocardiographic (ECG) changes in patients with partial epilepsy which two types ECG beats (partial epilepsy and normal). A comparison of the results shows that the performance of the evolutionary scheme outweighs that of support vector machine. In the best condition, the accuracy rate of the proposed approaches reaches 100% for specificity and 95.81% for sensitivity

کلیدواژه ها:

ANFIS ، Support vector machines ، Genetic algorithm ، Genetic-fuzzy SVM ، Partial seizure ، ECG