Neural Networks for Prediction of Loan Default Using Attribute Relevance Analysis

Predicting the class label using neural networks through attribute relevance analysis is presented in this paper. This method has the advantage that the number of units required can be reduced so that we can increase the speed of neural network technique for predicting the class label of the new tuples. In this proposed paper attribute relevance analysis is used to eliminate irrelevant attributes to give as inputs to neural network. A simple neural network is used for testing class defaulter. The results shows that this method is feasible.