In anticipation of his upcoming presentation at Predictive Analytics World Finance, part of Machine Learning Week, June 18-22, 2023 in Las Vegas, we asked Mack Wallace, Head of Financial Products at MPOWER Financing, a few
questions about his presentation, Machine Learning in Credit Decisioning: Using New Data & Methods to Approve Students with No Credit History — see what’s in store at the PAW Finance conference.
Q: In your work with predictive analytics, what behavior or outcome do your models predict?
A: At MPOWER, we are able to approve people for affordable student loans with no credit history. We do this through application of new data sources and sophisticated models that predict a prospective student’s future career prospects based on observable factors today – like academic performance, the quality of the program they are pursuing.
Q: How does machine learning deliver value at your organization – what is one specific way in which it actively drives decisions or operations?
A: In addition to underwriting, machine learning delivers value throughout our organization. Within credit decisioning, we use machine learning to safely and responsibly approve more people. We use OCR extraction to validate documents and automate business processes. We use churn and propensity models to assist in campaign targeting. We are also exploring how new advances in machine learning, like the LLMs that power ChatGBT, can be used to facilitate better, faster customer responses for their questions.
Q: Can you describe a quantitative result, such as the predictive lift of your model or the ROI of an analytics initiative?
A: As a result of better predictive lift, we have a stellar record for loan credit losses – often comparable or lower than market leaders in our industry with decades of experience.
Q: What surprising discovery or insight have you unearthed in your data?
A: Interestingly, we discovered that loan size and risk are often unrelated – that is, just because you borrow more, doesn’t also mean you are more risky.
Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Machine Learning Week.
A: As machine learning professionals, we must approach both how we frame the business problem and how we approach our modeling with equal intentionality.
Don’t miss Mack’s presentation — Click here to register for attendance.
You must be logged in to post a comment.
 |
 |
 |
Great interview. It is awesome like pizza tower
Through this interview, I can see Mack Wallace’s strategies in utilizing machine learning for credit decisions are clearly focused on leveraging new data sources and sophisticated models to make informed and responsible lending fnf decisions, ultimately benefiting both the organization and prospective students.
Machine learning delivers significant value across various organizations by optimizing processes, enhancing decision-making, and driving innovation. Here’s a specific example of how it can actively drive decisions or operations:
Example: Fraud Detection in Financial Services
Organization: Financial Services Company Melbet africa (e.g., a bank or payment service provider like PayPal)
Context:
In the financial industry, detecting and preventing fraudulent activities is crucial. Traditional rule-based systems can be slow to adapt to new fraud patterns, leading to potential losses and inefficiencies.