Multivariate time-series classification of sleep patterns using a hybrid deep learning architecture
Jeonghan Hong, Junho Yoon · 2017
With the growing public interest in health today, people are rapidly increasing their use of sleep sensing devices and smartphone apps in their daily lives to check and manage their sleeping health. However, some of the current sleep monitoring services are void of technical reliability in terms of data collection and analytic methodologies. In this research, for the purpose of robust representativeness, Internet-of-things (IoT) sensors were utilized for precise and sufficient data collection and a hybrid of Deep Belief Network (DBN) and Long Short-Term Memory (LSTM) was proposed for accurate sleep patterns classification. In addition, we explore people's sleep sequence clusters and examine differentiations between them.