Contextualized behavior patterns for change reasoning in Ambient Assisted Living: A formal model

Abstract

International audienceAmbient Assisted Living, smart environments and technology used for elder care, increases independent living time and cuts long-term care costs. An important requirement for these systems is detecting and informing about abnormal behavior in users' routines. Still, routines may change when certain context conditions hold, which does not mean these changes should concern caregivers. However, changes due to context conditions are not often considered in current proposals. In this paper, we formalize the main concepts related to activities in AAL and we introduce contextualized behavior patterns, a long-term behavior model that considers variability due to context. We adopt a semantic similarity that allows to better detect behavior changes and to understand what makes an observation of daily living different from expected patterns. The results of the experience with three public data sets are promising. Keywords: behavior analysis · behavioral patterns · activities of daily living · elder care · context awareness · long term behavior modelin

The ContextAct@A4H Real-Life Dataset of Daily-Living Activities

“…Context features are often not measured, being the ones measured temperature at some locations and presence in rooms. Nonetheless, rich context information can not only improve activity recognition but also it can help to better understand behavior patterns and user triggers (for example, loud noises may trigger anxiety) [13].…”

A Framework for Detecting and Analyzing Behavior Changes of Elderly People over Time Using Learning Techniques

“…The research areas covered by the CASAS project include assistive technology, artificial intelligence, machine learning, and activity recognition. The benchmark dataset described in reference [ 14 ] was also made available as part of this project and is widely used to analyze behavior of all types of people including elderly people [ 15 , 16 ]. One of the works developed in the CASAS project concerns the design of a lightweight smart home called “smart home in a box” [ 17 ].…”

“…During our experimentations, we used one of the public datasets (named HH120) [ 14 ] which was also used in other works like [ 16 ]. This dataset contains sensor data that was collected in the (smart) home of a volunteer adult.…”

Context-Enriched Regular Human Behavioral Pattern Detection From Body Sensors Data

“…However, how to discover such information from a temporal database is an open issue that must be resolved. Context-aware batch learning algorithms have also been proposed [21], [22], [70]. To sum up, the discovered challenges are as follows:…”