A Novel Optimized Classifier For the Loan Repayment Capability Prediction System

Abstract

The most suitable predictive modelling technique to predict the loan repayment capability of a customer in a banking industry is classification. Classification is a supervised learning technique in data mining. The loan repayment capability of a customer can be predicted more accurately using random forest algorithm. The accuracy of the prediction depends on various parameters of the random forest algorithm. The main objective of this paper is to prove that optimization of parameters results in a better accuracy for the capability prediction of loan repayment by the customers. This paper illustrates the process of optimization that leads to an improved accuracy in classification. The comparative study explains that optimization can lead to a better accuracy and the experiments were done in weka and R.