Granular Decision Tree and Evolutionary Neural SVM for Protein Secondary Structure Prediction

A new sliding window scheme is introduced with multiple windows to form the protein data for SVM. Two new tertiary classifiers are introduced; one of them makes use of support vector machines as neurons in neural network architecture and the other tertiary classifier is a granular decision tree based on granular computing, decision tree and SVM. Binary classifier using multiple windows is compared with single window scheme. The accuracy levels of the new classifiers are better than most available techniques.