A German direct bank faces the challenge of migrating the prediction of loan defaults into the new Big Data infrastructure. The prototype of a scoring model for predicting loan default probabilities from a previous project is to be implemented in Spark.
In the first step, several possible Big Data technologies are evaluated for the implementation. After the evaluation, the model is implemented using the combination of Spark and H20 (Sparkling Water) using a random regression forest.
The implemented model can be used within the new Big Data infrastructure for online scoring of customers. The implementation makes it possible to re-train the model with current data at any time and thus optimise the bank's default risks. The risk management requirement is also met by historising and storing the models.
Are you interested in your own use cases?
Our Case Studies
- Get even more detailed insights into our customer projects -
Smart cooking with the Thermomix
Data Operations at Munich Re
Data & AI Knowledge
Creating added value from data & AI together