Statistical modeling of the resident's activity interval in smart homes

Abstract

The activities of residents in smart homes possess temporal information which can be used to classify and model psychological behavior of the resident. In this study, a learning algorithm is proposed to predict the activity interval of smart home inhabitants. The algorithm is based on the hypothesis that residents' activity intervals follow a normal distribution. To predict the starting time of the following activity, it incrementally utilizes mean and standard deviation of previous history which are applied according to the central limit theory of statistical probability. The prediction algorithm exhibits 88.3 to 95.3% prediction accuracies for different ranges of mean and standard deviations when verified by practical smart home data. Further stochastic analyses prove that the time difference between the residents' activities follows normal distribution which was merely an assumption previously. © 2011 Asian Network for Scientific Information.