Performance Evaluation Analysis of MLP & DG-RBF Feed Forward Neural Networks for Pattern Classification of Handwritten English Curve Scripts

The purpose of this study is to evaluate the performance analysis of multilayer feed forward neural networks trained with back propagation algorithm & descent gradient Radial basis function network for the pattern classification of hand written curve script. This analysis has been done for handwritten text of three letters and for the individual English vowels. This analysis in the performance has been evaluated for the five different samples of handwritten English vowels and handwritten text of the three letters. Evaluation process is executed upon raw data in binary form and data based on extracted features (tangent values & density value) for each word & vowels. These characters are presented to the neural network for the training. Adjusting the connection strength and network parameters perform the training process in the neural network. The results of 3600 experiments indicate that the feed forward MLP performs accurately and exhaustively with imposed DG-RBF method. Key wordsPattern Classification, Feed forward neural network, Back propagation algorithm, Radial Basis Function Neural Network.