Recommender systems for IoT enabled quantified-self applications
Seda Polat Erdeniz, Andreas Menychtas, Ilias G. Maglogiannis, Alexander Felfernig, Thi Ngoc Trang Tran · Evolving Systems · 2019
Abstract As an emerging trend in big data science, applications based onthe Quantified-Self (QS)engage individuals in the self-tracking of any kind of biological, physical, behavioral, or environmental information as individuals or groups. There are new needs and opportunities for recommender systems to develop new models/approaches to support QS application users.Recommender systemscan help to more easily identify relevant artifacts for users and thus improve user experiences. Currently recommender systems are widely and effectively used in the e-commerce domain (e.g., online music services, online bookstores). Next-generation QS applications could include more recommender tools for assisting the users of QS systems based on their personal self-tracking data streams from wearable electronics, biosensors, mobile phones, genomic data, and cloud-based services. In this paper, we propose three new recommendation approaches for QS applications:Virtual Coach,Virtual Nurse, andVirtual Sleep Regulatorwhich help QS users to improve their health conditions.Virtual Coachworks like a real fitness coach to recommend personalized work-out plans whereasVirtual Nurseconsiders the medical history and health targets of a user to recommend a suitable physical activity plan.Virtual Sleep Regulatoris specifically designed for insomnia (a kind of sleep disorder) patients to improve their sleep quality with the help of recommended physical activity and sleep plans. We explain how these proposed recommender technologies can be applied on the basis of the collected QS data to create qualitative recommendations for user needs. We present example recommendation results ofVirtual Sleep Regulatoron the basis of the dataset from a real world QS application.