Towards a Theory of Space for Activity Recognition in Smart Environments Based on Rough Sets

This paper investigates how spatial information can be used to improve the process of recognising activities in smart homes. The spatial information is obtained through spatial markers, like motion sensors for instance. Since different markers may be involved in different occurrences of the same type of behaviour, there is a need for being able to express vagueness, which in our approach is achieved by using rough sets rather than standard sets to denote the space in which an activity occurs.