The Challenge:
To create better models for credit card default prediction among current clients, without needing to utilize any additional features other than the ones already available.
The Solution:
We used deep learning models to encode temporal features that classic approaches could not facilitate during optimization.
The Outcome:
We created a more accurate representation of customers who would been using the product for more than 6 months (50% of users), and reduced the effort required for feature engineering.
Tech Tools & Skills Applied:
We used a variety of skills and tech stack tools including: TensorFlow, Python, Deep Learning, RNNs, Forecasting, and Credit Scoring.
Results
Stuff
Stuff
