An Analysis of Case Study on Application of Backpropagation Network Using Procedural & Object-Oriented Approach

The backpropagation (BPN) algorithm is used in artificial neural networks (ANN) for training the networks that uses unsupervised learning and it is performed in two consecutive phases for the training. In the first phase, a pattern is provided as input of network and in response to that activities are performed layers-to-layers on inner layers till a response pattern is generated at the output layer. The second phase basically compares the response pattern generated with the input pattern provided and finds the mismatch that occurred. These mismatches for the target values define the error values and are backpropagated to the input layer as feedback. On the basis of this feedback, the weights for inner layers are adjusted again and again in a loop of backpropagation, such that the final error can be minimized. This work has compared and contrasted the recent behavior between training neural networks using the backpropagation algorithm using a structured & object-oriented approach.