CRM (credit risk modeling)
To lend a loan is an exhaustive process. Hence, there should be an automated way to predict whether the applicant will be paying off the loan or be written off as default. Using KNIME, it becomes easy to do predictions. On the same foundation, we proposed a solution. We predicted whether the applicant will default an installment or not. Some factors like monthly debt, credit history, number of bank accounts and their details, bankruptcies, and tax liens were taken into account for it. Therefore making this project is a boon to the financial sector. After understanding the Dataset, the most important thing is to clean the data and visualize them with some graphical representation so it helps to analyze the data and make some conclusions from the data. We are solving this problem using the KNIME Analytics Platform, it’s a No-code Analytics platform so for this particular approach we are using KNIME nodes for each process.
The accuracy obtained at the end of the implementation process was 82%.