Professor Glen Urban
Sponsor: Suruga Bank
Deep Learning is one of the most exciting new technologies in artificial intelligence. It is currently used for voice recognition and image identification in firms like Google, Facebook, and Apple. Most recently deep learning algorithms were used to create a program that beat one the world’s best GO players. The purpose of this project is to apply this technology to the development of new products in Financial Services. The impact is initially at the opportunity identification phase where the area for potential new products/services are identified. We propose to use big data (customer demographics, site level clicks, current product ownership, consumption records) to parameterize a deep learning model that can simulate the likely response to new product/service configurations (e.g. new credit card with cash rewards, moderate interest, zero interest on balance transfers for 6 months, and a high borrowing limit, versus a card with high travel rewards, high interest rates, regular interest rates on balance transfers, and mid borrowing limit). This model will allow virtual testing of new product/service configurations. If an attractive configuration is identified, then this opportunity can be tested by concept and pre-test market analytical procedures. This represents a cost-effective methodology of developing customized financial services products that better serve customer needs.