 |
Home Human behaviour monitoring based on recorded data from existing smart thermostat in North American houses
1 : Université de technologie de Compiègne, CNRS, Biomechanics and Bioengineering, Centre de recherche Royallieu – Compiègne, France
utc
2 : University of Waterloo [Waterloo]
3 : Université de technologie de Compiègne, CNRS, Biomechanics and Bioengineering, Centre de recherche Royallieu – Compiègne, France
utc
In the context of increasing life expectation and the wish of elderly people to life home, monitoring system able to identify activities are needed. In order to develop this type of system large data base are needed. A possible solution is to use information from existing sensors used for other applications. In this paper we propose an approach of machine learning algorithms applied on a large data from Ecobee sensors in many houses in North America. The first results of classification are encouraging.
