Abstract:
Loans account for a large portion of bank profits. Despite the fact that many people are
looking for loans. Finding a legitimate applicant who will return the loan is difficult.
Choosing a real applicant may be difficult if the process is done manually. As a result, we
are creating a machine learning-based loan prediction system that will choose the
qualified applicants on its own. Both the applicant and the bank staff will benefit from
this. There will be a significant reduction in the loan sanctioning period of time. In this
research. The majority of the bank's revenue is generated directly from the interest
income on loans. Even when the bank authorizes the loan following a lengthy verification
and testimony process, there is no guarantee that the chosen hopeful is the best hopeful.
When performed manually, this operation requires additional time. We have the ability to
foretell whether a specific hopeful is secure or not, and machine literacy has mechanized
the entire testifying procedure. Loan Prognostic is extremely beneficial for both banks'
clients and potential borrowers. I use some machine learning algorithms techniques to
predict the loan data. Those are the Decision Tree, K Nearest Neighbour, SVM, Naive
Bayes, and Random Forest Classifier. Nevertheless, I did uncover promising setups for
both purposes. I got the best accuracy from Naïve Bayes which was 82.16%.
