My main research is currently in the area of digital health. The specific research topics – within digital health and beyond – are:
Behaviour Change Support Systems:
A current research area is dealing with how to support people to change their behaviour to improve their health. To this end we have developed the SABA framework which facilitates the development of behaviour change support systems. SABA comprises components to accommodate user preferences and to adapt system interventions to individual users. User adaptation is realized by means of user modeling and collaborative filtering, resulting in a self-learning application that changes in line with a user’s progress and takes a user’s current situation into account. We expect our approach to enhance user acceptance and increase and sustain people’s motivation for behavioral change.
So far we have applied behaviour change support systems for reducing stress, improving sleep, and helping with increasing physical activity.
Making Sense of Sensors:
With the proliferation of body sensors on the market it has become quite easy to collect a wide variety of vital data as well as behavioural data indicating physical exercise. To make sense of all the collected data we employ data mining techniques to derive insights which then again lead to personalised advice, e.g. on how to change one’s behavior to improve sleep quality, or how to avoid stress. Since giving personalised advice can contribute considerably to behavioural change, sensor data analysis combines nicely with behavioural change support systems.
Application of Semantic Web technologies to various application areas such as knowledge management, e-health, e-government, and e-participation, for example to support personalised information access for patients, to enable clustering and categorisation of contributions to discussion forums, and extracting the key terminology from a text collection.
Domain-Specific Modelling Languages (DSLs):
Domain-specific modelling languages (DSLs) are focused on supporting the creation of models within a certain application domain. They typically comprise predefined domain-specific concepts as well as specialised language constructs that make it easier to create a model in the associated domain. In our current research we are applying DSLs and meta-modelling in the application domain of e-health, e.g. to help developing and configuring toolboxes of IT-based components for various medical interventions.