Page - 300 - in Intelligent Environments 2019 - Workshop Proceedings of the 15th International Conference on Intelligent Environments

The second term, “independent living” is definedas “ahousing andcommunity ar- rangement thatmaximizes independence and self-determination”. Both termswere in- troduced in2003and2010 respectively. It isworthnoting that the terms: living independent, communitydwelling, andage- ing inplace, arealsoused inhealthcareas synonyms for independent living. 4.3. SmartHouse According to a review on smart houses welfare technology, “Smart house commonly refers toany livingorworkingenvironment carefullydesigned toassist residents incar- ryingoutdailyactivitiesand topromote independent lifestyles” [7,10].Thegeneralgoal is toadapt thehouse to theoccupant’spreferences. When searching inMeSH , there are two terms found.Thefirst term is from1992, “housing for the elderly”, defined as “housing arrangements for the elderly or aged, in- tended to foster independent living. The housingmay take the form of group homes or small apartments. The concept includes housing for the elderlywith some physical limitations”. Thesecondterm is from1968,“homesfor theaged”,definedas“geriatric long-term care facilitieswhichprovide supervision andassistance in activities of daily livingwith medical andnursingserviceswhen required”. Synonymsfoundinthehealthcaresciencesforhousingfortheelderlyare:retirement life care centers, continuing care retirement centers. In the sameway, the synonym for homes for theaged isoldagehomes. 4.4. ActivityRecognition Activity recognitionhas been implemented for several decades [4]. People usually tend tofollowapattern in theirdaily life[33,34].Hence, recognizingthepatternof theperson helps toadapt a smart house to the inhabitant. Commonly, theactions recognizedareactivitiesofdalily life (ADL), suchas sleep- ing, toileting,showering,dressing,eating,etc.Theseareactions that require“basicskills and focus on activities to take care of one’s ownbody” [35]. Byfinding repetitive pat- terns on the person’s activity, it is possible to predict the next activity of the person for assistance ifneeded [5]. 4.5. BehaviourModelling There is a few studies on behaviourmodeling as described in section 2. Yet, the term behaviour has not been fully defined as the previous terms presented in this section.At USN, behaviour is defined as the activity, duration, posture, and location of the person. E.g., havingbreakfast.Thus, abehaviour is anactivitywithduration [8,9]. HBMhelps to predict the next behaviour of the person and thus adapt better the smart home to its resident aswell as construct a safer environment. In addition,mod- elling the behaviour of the person allows detecting any anomaly on the person’s daily pattern. “Anomalydetection refers to theproblemoffindingpatterns indata that donot conform to expected behaviour” [36].Detecting anomalies can help alert familymem- bersor caretakers incaseofdangerous situations suchas falls. V.G.Sanchez /WelfareTechnology,Healthcare,andBehaviourModelling–AnAnalysis300