Credit scoring using Artificial Immune System algorithms: A comparative study

Abstract

Credit scoring has become a very important task in the credit industry and its use has increased at a phenomenal speed through the mass issue of credit cards since the 1960s. Credit scoring models have been widely studied in the areas of statistics, machine learning, and artificial intelligence (AI). Many novel approaches such as artificial neural networks (ANNs), rough sets, or decision trees have been proposed to increase the accuracy of credit scoring models. artificial immune systems (AIS) which are algorithm developed with inspiration from natural immune system processes have been used to solve various kinds of real life processes with success. Various AIS algorithms like AIRS, CLONALG, Immunos etc have been proposed. This paper explores the possibility of application of various artificial immune system algorithms to credit scoring problem and compares the results with other methodologies. Experiments are done against two benchmark data sets and results presented with respect to other algorithms to help credit analysts chose from various methodologies.