Smart home resident identification based on behavioral patterns using ambient sensors

摘要: In this paper, a novel approach is presented to identify the smart home residents. The different behavioral patterns of home’s inhabitants are exploited distinguish variation specific individual behavior in homes significant challenge. We introduce features that useful handle problem. Moreover, we an innovative strategy which considers Bag Sensor Events and Bayesian networks. approach, frequency each sensor event occurrence considered, regardless order events. efficiency compared Sequence Events. Our experiments confirm outperformed previous approaches. When residents people who repeat their daily activities frequently, applying on Activity Based Window Frame features, performed activities, would them more accurately. contrast, cases where perform ways, considering Time leads higher accuracy distinguishing created by constant time intervals. F-measure our proposed Twor2009, Tulum2009, Tulum2010 datasets 96%, 100%, 99%, respectively, improves results researches consider