Ministry of Electronics and Information Technology (MeitY) is testing credit rating model, a new system that will assist banks in assessing credit risk and probability of fraud using big data analysis. The system is expected to help banks, particularly rural and cooperative banks, tackle the issue of rising nonperforming assets (NPAs). Currently, rural and cooperative banks depend on judgement of bank manager, resulting in high NPAs and frauds. A credit rating model can assist banks to quantify risks under the big data context.
The MeitY sponsored project includes Reserve Bank of India (RBI), Bangalore-based IT firm Processware System and two cooperative banks as partners. The project is aimed at helping banks quantify risks associated with retail loans such as personal loans, gold loans and vehicle loans.
Under this project, a statistical and machine learning algorithmic model has been developed to predict probability of default with aim to reduce NPAs. It will also help in predicting different types of frauds in banking sector based on RBI guidelines. Besides, a web-enabled software is also being tested to assist banks to easily adopt models for credit rating, NPAs and fraud. Validation of these models has been done using data from several banks.