Cheerful status prediction using "Sleep Diary" application

Abstract

Recently, the problem of insomnia is a social issue. To solve this problem, we develop “Sleep Diary” application depend on the theory of cognitive behavioral therapy for insomnia: CBT-I. It is suggested that users continuously use this application may improve their sleep habituation. However, the continuation of behavior change is difficult for many people. So, we consider that it is required extrinsic motivation method to continuously use the application. In this study, we focused on the cheerful status prediction function as extrinsic motivation method. As a result, our method using random forest classified cheerful status with practical accuracy. Particularly effective features were chosen from average and standard deviation of 7 days sleep data. We will confirm effectiveness of this method.