AN EXPLORATORY STUDY BASED ANALYSIS ON LOAN PREDICTION

Abstract

With the advancement in the financial sector, more people are seeking for bank loans. However, banks have limited resources that they must allocate to certain persons, analyzing and evaluating as to who would be a potential risk to the bank and who would not be, determining the credit to be given to the risk-free consumer. So, this research is an attempting to reduce the risk element associated with selecting the protected individual in order to save a large number of bank asset, endeavors and resources. It's done by trawling through previous records of those who have received advances previously granted, and the machine was constructed based on these records using the AI model that provides the most exact result. This research deals with the motive whether or not it is safe to lend a loan to a consumer keeping in mind the amount should be returned on time and the credit is safeguarded or not. It is one of the most pressing and significant factors related to the financial institutions and equivalent businesses, since it has a considerable influence on their net revenue and profitability. The presence of multi non-performing mortgages has risen considerably in recent years, jeopardizing the growth of these Banking Institutions. We present a method for implementing a neural nets model that will be used to forecast and predict loan mortgage default. The projection is made by considering the personal and monetary information given by the probable loan taker. The data-set which is used to train and evaluate the suggested neural network model. Our suggested neural network model surpasses alternative classifiers that are usually employed by monetary firms for loan default forecasting, based on the findings obtained. The accuracy obtained by our model on data set is close to 97.8%.