Credit analysis using radial basis function networks

The assessment of credit applications is usually carried out by loan officers based on their own heuristics. Thus, different officers may make different decisions for the same application. In order to avoid the problems due to this subjective evaluation, quantitative methods have been proposed, such as the use of neural networks. In this paper, different approaches to define the parameters of radial basis function (RBF) neural networks are applied in a credit evaluation task.