Nudging and Behavioural Machine Learning: Implications for Urban Wellbeing

This research will focus on designing and building behavioural 'nudges' using insights from the Behavioural Machine Learning methodology and testing them using field experiments. The student will use data inference and machine learning algorithms to develop ways in which end users (citizens, households, policy-makers, and businesses) can change their behaviour in order to reach higher levels of urban wellbeing. He will then design and conduct laboratory and field experiment to test how wellbeing data and behavioural insights should be presented to citizens, households, policy-makers, and businesses in order to 'nudge' these actors to make more optimal decisions and to increase urban wellbeing. The project will focus on the design and development of: (i) feedback protocols and (ii) incentive mechanisms, informed by the work of the project team on BML, HCI/HDI; and (iii) evaluation of mechanisms for specifying happiness requirements as well as determining the data flow between the platforms used by the team, partners (SunBath, HAT, and Databox), and external applications.