SHMUC prototype > F_Context_Switching_Use_Cases
As discussed above, context is one of the most important factors that impact on human behaviour. Context varies from data about the environment such as temperature and moisture level, to the temporal and spatial data: season, day of week, etc., through to to other simultaneous activities. Under different contexts, people will behave in different ways. Therefore, a smart home system should have different behaviour models for different contexts and be able to choose a model suitable for the current context, so that it can make a better classification.
F1 The Working Week
This is an interesting aspect of the smart home application, especially for human activity recognition and abnormal behaviour detection, since both them depend strongly upon the context. It represents the adaptive ability of the system to the change of context, and therefore plays an important role in the accuracy of recognition and detection. Moreover, switching between states raises the idea of building an ubiquitous smart home system that can aggregate data from many other sources, e.g., from many different smart homes, and use it inheritably. For example, suppose that each house could inform others when their inhabitant was ill. This could lead to an early warning system for pandemics, where other houses warn their inhabitants to wash their hands frequently and stock up on food in case they become ill.