Solving simple daily tasks is usually accomplished automatically. However, this might present a hard struggle for elderly people and individuals with disabilties or cognitive impairments. Ambient Intelligence and ubiquitous information technologies are promising perspectives for such systems, particularly what David L. Tennenhouse of Intel calls ”proactive computing” and which suggests an anticipatory aspect to ubiquitous computing.

According to Robert Rosen, an anticipatory system can be defined as ”a system containing a predictive model of itself and/or of its environment that allows it to change state at an instant in accord with the model’s predictions pertaining to a later instant”. Generating such predictive models necessitates context awareness. This provides the means to determine what to observe and how to interpret these observations, which, in turn, helps improve this predictive model of the real world and then triggers the corresponding changes on the state of the world.

In my research, I address some of the underlying principles of anomalies preventive systems for proactive applications in closed domains based on activity recognition.