Towards an understanding classification of well-being for care of older people

As the ageing population increases, more and more people require care and support. To date there has been limited understanding of older people´s needs and the general well-being when they are in care. In this paper we investigated the relationships between the indices of independence in daily activities of the people who are in care and the behaviour rating scale [2] defined in the nursing assessment in connection with the indices to predetermine individuals´ well-being during their stay. For this study, we focus on three well-being attributes, Cognition, Social Relations and Communications which are important for understanding the people in care. The study is based on a set of datasets consists of 40,000 records. The datasets were trained, tested and analysed using four Machine Learning (ML) classifications algorithms that are Bayesian Networks (BN), Naïve Bayesian (NB), Naïve Bayesian Tree (NBTree) and Instance Based Learner (IBL). The results show that Bayesian based algorithms performs very good with well-being data with respect to determining the relationship between the daily activities and the behavior rating scale, to predetermine the needs and well-being of the people who are in care.