About the client
The client is a leading community-based financial services company in San Francisco. They offer a wide range of financial products and services to both individuals and businesses, including banking, investments, mortgages, and consumer and commercial loans.
Client challenges
- The client wanted to automate the loan application eligibility review process that slowed down decision-making and approval workflows.
- Eliminate manual data preparation for monthly reports while eradicating data quality issues due to a lack of an efficient cleansing framework.
- Develop a predictive model with high accuracy for determining qualification statuses such as credit score and personal history is absent.
Solution
- We designed eligibility prediction models and balanced them using the Synthetic Minority Oversampling Technique (SMOTE).
- Developed ML models using Gradient Boosting, Logistic Regression, Random Forest, and XGB Classifier technologies.
- Improved the performance of the ML model using hyperparameter tuning, which was measured and evaluated through Accuracy and Area Under ROC Curve (AUC) metrics.
- Power BI implementation provided valuable insight into loans status (approved/rejected) by reason.
Benefits
- ML model replaced the manual loan application solution.
- Automated approval workflow improved decision-making.
- Reduced the decision-making time w.r.t loan eligibility from days to minutes.
- Automated data cleansing and preparation for monthly reporting using Power BI.
- Ability to visually present loan status by reason using Power BI.