A genetic algorithm-based deep RBF neural network for medical classification

摘要: In this paper, we introduce a deep RBF neural network for medical classification. The proposed classifier consists of two parts: an auto-encoder and network. is used to decrease the number characteristics presented samples. Then, obtained new features are design networks performed in stages. First, subtractive clustering method define centers RBFs. Second, genetic algorithm optimize widths To assess classifier, perform tests over three datasets from UCI machine-learning repository compare its performances with other methods.