• This work explores how deep learning can be used to classify mortgages into Default or Fully Paid Loans.
• The research uses the Freddy Mac Single-Family Loans dataset, which is used to create novel geographically based features on a per loan basis.
• The model is trained on animbalanced dataset which includes using 15.3 million unique loans with 326 million performance updates, achieving an AUC score of 0.984.

Acedemic paper coming soon ... ;)