Behavioral analysis, based on unobtrusive monitoring through environmental sensors, is expected to increase health awareness of AAL systems. In this paper, techniques for assessing behavioral quantitative features are discussed, suitable for detecting behavioral anomalies in an unsupervised fashion, i.e., with no need of defining target reference behaviors and of tuning user-specific threshold parameters. Such technique is being exploited for analyzing data coming from a set of European pilot sites, in the framework of the EU/AAL-JP project 'FOOD', specifically focused at kitchen activity. Simple results are illustrated, suitable for proof-of-concept validation.
Self-tuning behavioral analysis in AAL 'FOOD' project pilot environments / Mora, Niccolo'; Losardo, Agostino; DE MUNARI, Ilaria; Ciampolini, Paolo. - 217:(2015), pp. 295-299. [10.3233/978-1-61499-566-1-295]
Self-tuning behavioral analysis in AAL 'FOOD' project pilot environments
MORA, Niccolo';LOSARDO, Agostino;DE MUNARI, Ilaria;CIAMPOLINI, Paolo
2015-01-01
Abstract
Behavioral analysis, based on unobtrusive monitoring through environmental sensors, is expected to increase health awareness of AAL systems. In this paper, techniques for assessing behavioral quantitative features are discussed, suitable for detecting behavioral anomalies in an unsupervised fashion, i.e., with no need of defining target reference behaviors and of tuning user-specific threshold parameters. Such technique is being exploited for analyzing data coming from a set of European pilot sites, in the framework of the EU/AAL-JP project 'FOOD', specifically focused at kitchen activity. Simple results are illustrated, suitable for proof-of-concept validation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
