Yen, Y.C.Y.C.YenLu, C.H.C.H.LuCheng, Y.C.Y.C.ChengChen, J.S.J.S.ChenLI-CHEN FU2020-05-042020-05-042011https://scholars.lib.ntu.edu.tw/handle/123456789/488905Due to the rapid growth of the aging population, numerous countries have been attaching importance to establishing the well-being of the elderly. However, long-term healthcare is labor intensive. To alleviate the possible social costs associated with manpower and physical resources, we propose an evidence-based caring system which can inconspicuously and automatically monitor the health status of the elderly by continuously analyzing their real-life long-term living patterns deduced from activity recognition. In this way, caregivers can get hold of the behavior changes even the elderly is not under caregivers' supervision. Moreover, we adopt a persuasive policy to provide timely reminders and encourage the elderly to achieve a healthier life. In the primary stage, we do preliminary experiments in a nursing room. Based on the experiment, we conduct several interviews aiming to improve our system in the next phase. ? 2011 Springer-Verlag.[SDGs]SDG3Activity recognition; Aging population; Behavior change; Caring system; Context-Aware; elderly healthcare; Health status; Labor intensive; Living pattern; Persuasive technology; Physical resources; Rapid growth; Social cost; Behavioral research; Experiments; Health care; Human computer interaction; Knowledge management; Population statistics; Technology; Pattern recognition systemsconference paper10.1007/978-3-642-21666-4_27https://www.scopus.com/inward/record.uri?eid=2-s2.0-79960324840&doi=10.1007%2f978-3-642-21666-4_27&partnerID=40&md5=f0ebecbdfd183dbb23bd69edd1eb9126