Abstract: India has been ranked 5th in the list of countries with the highest non-Performing Assets. It implies India possesses more defaulters each year regarding the subject’s ability to repay their loans. Noticeably this is an alarming figure and with the rise of Bank frauds, figured at high as 100 Crore per day. With this as our motivation, we’ve decided to work on a problem that suitably allows us to identify whether or not a bank can engage in business with a client based on their previous history. Using ML techniques taught, our project aims at creating an efficient model where we can appropriately find the right set of metrics from a diverse group of features that help us in identifying whether a subject can repay their loan once taken.

Keywords: Banks, loans, repay, Non-performing assets, frauds, defaulters

| DOI: 10.17148/IARJSET.2022.94122