Personalisation is the problem of tailoring products, services, or communications to individual client preferences at scale, where a client might be any person interested in a business or organisation's services. For example, marketing can be better targeted if brand preferences are accurately understood. Similarly, search results can be filtered and ordered according to individual needs. In this project we study a new approach to personalisation which addresses the ‘cold start’ problem. This is the problem of how to personalise when a new client arrives and very little is known about them.
The approach is based on a novel model of human decision making that permits advanced machine learning algorithms to infer preferences from data more quickly and accurately than existing approaches. The project investigates the feasibility and market potential of the new personalisation approach.