ABSTRACT
In old age, a series of common health conditions, chronic diseases, and disabilities affect the individual's physical and mental health and prevent the performing of Activities of Daily Living. This paper presents a solution to identify abnormalities in the behavior of the elderly based on ADL (Activities of Daily Living), using Machine Learning, through the Novelty Detection technique. The ADL data were used to create a model that defines the baseline behavior of the elderly, and new observations, to verify significant changes in behavior, are classified as discrepant or abnormal. The Local Outlier Factor, One-class Vector Machine, Robust Covariance, and Isolation Forest novelty detection algorithms were used and evaluated. The model presented reached an accuracy and F1-Score of 96%.