GMDH polynomial and RBF neural network for oral cancer classification

This paper studies the group method of data handling (GMDH) polynomial neural network and radial basis function neural network (RBF) for classification of oral cancer. The oral cancer dataset of 1025 patients is divided into two subgroups: training data and test data, to verify the network’s ability to diagnose new cases. Classification accuracy of the GMDH model for training data is 70.24 % and that for the validation data is 67.80 %, whereas the classification accuracy of the RBF model for training data is 71.02 % and that for the validation data is 67.71 %. The performance of both the neural network models is found to have competitive results; thus making both the models efficient for classification.