A Case Study on the Analysis of Behavior Patterns and Pattern Changes in Smart Environments

Abstract

Abstract. Societies need to devise mechanisms of caring for the well aging of the increasing number of seniors, as it is very important for elderly people to maintain their independence. Smart environments are being devised as a form of care in what has been called ambient assisted living. A smart environment should be able to respond in case of emergency or risk and inform any abnormal behavior. Still, not much research is done to understand behavior patterns, temporal changes and other particularities that can affect the effectiveness of smart environments in ambient assisted living. We explored the behavior of two adults in a smart environment in order to reveal temporal, spatial and sequential relations among the activities as well as the changes that these relations undergo overtime and across individuals. This paper presents an analysis of three human behavior patterns: temporal, location and frequency. These patterns are mined on two experimental subjects using the dataset provided by the CASAS project.Our analysis evidences how temporal, spatial and sequential patterns differ from person to person, day to day and after some time. Learning personalized behaviors and identifying and adapting to changes is a crucial aspect for smart environments since one-fit-all solutions are not suitable.

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

“…-Intra-class variability: the same activity can be performed in different ways, using different objects, at different locations, etc. [14,15]. -The Other activity: scripted datasets only represent activities of interest, while in real-life scenarios most of the time it is other activity is being performed [12].…”

“…Nevertheless, ontologies lack support for the temporal characterization of activities [19] which is an important aspect of activity recognition as they take place during a span of time. For ADL recognition it is important to consider context information such as environmental conditions, visitor presence, current location and part of day [14]. Nonetheless, the usefulness of different context variables has not been proven in real-life scenarios due to the lack of available data in the same dataset.…”

“…ContextAct@A4H includes sensor raw data reflecting the context as well as standard daily living annotated activities. Context information can help study how contextual changes affect user behavior and preferences [14]. To the best of our knowledge it will be the first public available dataset including such rich real-life data.…”

Representing and Learning Human Behavior Patterns with Contextual Variability

“…The usual start time of a behavior is modeled as the interval(s) with the highest observation frequencies or as [µ−σ i , µ+σ]. But these times vary according to context conditions without them being anomalies [8]. To model these variations some authors create a different model for each possible situation [9,10].…”