Internet of thing has been one of the greatest advances in our digital era, with this small computers we can make our lifes easier that never before. However one of the biggest concerns about this technology is that they tipically rely on cameras and microphones to provide us useful information.
Our objetive consist in a non intrusive way (without the use of cameras or microphones) of detecting if a room is occupied or not. To address this problem we will use the 'Occupancy Detection Data Set', free and open to the public on the web portal of the UCI. This database consists of a set of observations made by sensors and then checked using photographs taken every minute to discern the occupancy and values of sensors that were considered relevant. So a human factor provided the clasificacíon through photographs and it is intended to use environmental sensors to achieve the same goal but reduce intrusion.
The data set used for this study was Occupancy Detection Data Set
This study is developed as part of the Machine learning course of the University of Granada that is the reason why it is in Spanish. Nevertheless we intend to translate it into English soon.