Prediction of Loan Status in Commercial Bank using Machine Learning Classifier

Abstract

To address a range of difficulties in the banking business, the
creation of a more accurate predictive modelling system is
required. It is impossible to forecast who would fail on a loan,
which makes it an onerous job for the banking industry. When i t
comes to loans, one of the quality factors is how they are
currently doing. Despite the fact that it does not show a l l o f the
information right away, it is the first stage in the loan application
process and must be completed. The loan status is taken into
account while developing a credit rating model. Credit scoring
algorithms are utilised for trustworthy credit data analysis, since
they allow for the identification of defaulters and legitimate
customers in the credit data. One of the objectives of this proje ct
is to create a credit scoring model that is based on credit
information. For the purpose of developing the financial cred i t
score model, a number of machine learning algorithms are used
in tandem with one another. This study includes the development
of a credit data analysis approach that is based on machine
learning classifiers as part of the overall project. KNearest
Neighbor (K-NN) classifier is used in combination with Min-
Max normalisation to provide the best results. The goal is
realised via the usage of the R programming language as well a s
other resources. In terms of the most important facts, the
proposed model provides the most accurate information
conceivable. Machine learning classifiers are used by
commercial banks to forecast the status of loan
applications.Keywords: Credit Scoring; K-NN; Loan status;
Loan Lending Process;Min-Max Normaliz