Um sistema de recomendação de atividades físicas para hipertensos no contexto de E-health

Abstract

Recommender systems are applied in a number of areas, including health in so-called e-Health systems. Most studies focus on nutrition and physical activity recommendations, with one of the goals to improve lifestyle. The purpose of this study were to develop a system for recommending physical activities for hypertensive patients, in order to facilitate the search for a healthy lifestyle. To achieve this goal, a hypertensive user profile model, called HyperModel2PAR, were developed and a hypertensive physical activity recommender system, called Hyper-RecSysPA. The resulting model is composed of 32 elements divided in three groups, used in the modeling of user profiles within the system of generation of recommendations Hyper-RecSysPA. Three cardiologists validated this system at different times. As a result, more than 75% of the recommendations generated were approved. Therefore, it is understood that the objectives of this work were achieved, since Hyper-Model2PAR and Hyper-RecSysPA obtained positive results in evaluations performed by expert users.