How DNB's Services Can Be Recommended To Their Customers?
Abstract
Checking today's weather might be the first thing you do when you wake up, followed by reading the news on the metro to work, finding out where to eat for the evening through restaurant recommendations, and playing Chess against other people before falling asleep. There is a continues growth of people interacting with web applications using their devices to stay updated and informed, social and entertained. All these actions are captured. The sites visited, Geo-locations, social network and so on, which forms an Internet profile for each user. This profile represents the user's habits and interests. Google, Amazon, and many other companies take advantage of these information-rich profiles to enhance and personalize their services.
In this thesis, we are investigating how banks can recommend their services to their customers, through personalization. This is done by generating a user profile model inspired by existing recommendation systems.
Limitations with respect to the dataset, and challenges with communications have led to a poor evaluation of the feedback. Our main conclusion from the evaluation is that this method has a potential, but more research is needed to make it work. Overall, we believe the theory of using a user profile model in banking will help DNB in the direction of personalization of their services.