Credit Risk Modelling for a leading SEA Loan Provider
The client has access to all loan and customer-related information in PDF format. This information can be obtained by entering a unique identification number associated with each customer. The project leveraged teX.ai to extract and summarize the data from the PDF files at a row level. Following this, predictive analytics was performed to evaluate the creditworthiness and loan eligibility of the customer. Power BI was used to create dashboards to provide an overview of the types of loans, repayment rates, customer churn rate, sales rep performance, and much more.
The client is a leading holding conglomerate that capitalizes on fintech, as well as the best emerging technologies in the market, to provide financial services to the underserved in Southeast Asia. The client owns the outstanding stocks of other companies and has access to loan-related information of financial organizations.
For each customer, there were two separate PDFs that needed to be merged, containing the following information:
- Loan-related details of the customer.
- Geographic information of each customer.
Since a customer could have taken multiple loans, it was necessary to summarize the information from these PDFs at a row level using specific business logic. By merging the loan-related data and the geographic details, a comprehensive view of each customer’s loan history and their corresponding geographic information could be obtained. This merged data would serve as a basis for further analysis and decision-making processes.