Prediction of residents' behavior using data mining of environmental information

Human lifestyles are evolving along with the advancement of electric devices such as computers and internet appliances. This evolution has brought diversity of our needs for living spaces in buildings. Human reactions associated with an event are different depending on individuals due to their knowledge on devices and their living preferences. Thus the services satisfying all of them are difficult to define. Therefore, a living space customized for each person is demanded. To achieve such customized spaces, it is important to recognize the human preference because it becomes possible to predict the human behavior if we can know this information. This paper proposes a method to predict residents' behavior using data mining of environmental information relevant to the living space.