Master Thesis
ALAS. Intelligent Anomaly Notification System based on Behavioral Daily Living Activities Patterns
Single-person households are increasing around the world, which is why many people find themselves in a situation of fragility when they deal with unforeseen events. For example, illness, accident or incidents in their homes. This work presents a notification model for anomalies on daily life behavior pattern activities, which are inferred from ambient non-intrusive sensors data analysis. The notification model considers a pervasive, non-intrusive, customizable and adaptable way of reacting, always guaranteeing the privacy of the user. Including the personalization of a support network of the inhabitant, analysis of sensor dataflows, adaptation to the interaction of the user with the system and non-intrusive feed-back.