SleepQual and B.Health: Smartwatch and Smartphone based Behavioral Datasets of Youth

Wearable sensors recording different components of peoples’ daily activity are commonly used these days for collecting motion data. These provide objective measures of behavioral components like sleep and physical activity. Smartphones are being used increasingly contributing to users’ screen viewing and hence are capable of providing precise measures of users’ screen time. Disruption in the key lifestyle patterns like sleep, physical activity and screen viewing may result in health impairments as these correspond to the utmost engaged times during the day. This study presents a novel dataset containing attributes of physical activity, sleep, and smartphone-based screen viewing collected from users’ smartphones and smartwatches. The dataset contains real time objective measures of activities of 24 undergraduate students. A sleep quality indicator and a behavioral health indicator namely SleepQual and B.Health respectively are evaluated using the extracted features. The instances are labelled based on the scores of SleepQual and B.Health. We have made the dataset publicly available which opens avenues for researchers to employ various intelligent data analysis techniques in order to develop systems capable of automatically assessing sleep quality and behavioral health using digital data without the need of any self-reporting questionnaires and tools. To provide baseline performance four machine learning classifiers are implemented on the proposed dataset.