A Prototype Artificial Neural Network to Assist in Credit Card Ii Applications Assessment at Visa Card

Abstract

This paper describes the construction of a prototype Artificial Neural Network to assist in credit card applications assessment for VISA card at the National Bank.

,The Artificial Neural Network was constructed using a neural network development \'tool, InfiNet, implemented in language C. Using the Backpropagation learning rule the network was able to learn from vector patterns it was exposed to.

The project report is divided into five sections. Section One introduces the use and management of credit cards and hence the need to perform assessment on applications before issuing a credit card. Section Two is dedicated to literature review. The objective of the literature review is to give an overview of Neural Network theory which a number of readers may not be familiar with.

Section Three describes the tools used in the development process and data collection. Section Four discusses howthe prototype Artificial Neural Network was designed and developed. It specifies all the ANN parameters. Section Five closes the discussion.

One definite conclusion emerging from this work is that Neural Networks are a useful tool in modelling expert decision making without having to perform the very challenging and time consuming task of knowledge elucidation required to build an Expert system. However, unlike Expert Systems they do cannot explain their