作者: Francisco E Martínez-Pérez , Jose Ángel González-Fraga , Juan C Cuevas-Tello , Marcela D Rodríguez , None
DOI: 10.3390/S120101072
关键词:
摘要: Human activity inference is not a simple process due to distinct ways of performing it. Our proposal presents the SCAN framework for inference. divided into three modules: (1) artifact recognition, (2) inference, and (3) representation, integrating important elements Ambient Intelligence (AmI) (artifact-behavior modeling, event interpretation context extraction). The extends roaming beat (RB) concept by obtaining representation using kinds technologies RB based on both analysis recognition from behavior A practical case shown in nursing home where system affording 91.35% effectiveness was implemented situ. Three examples are representation. Framework description, description CALog overcome problems such as feasibility implement AmI systems, show accomplishing challenges related recognition. We discuss how use RBs might positively impact faced designers developers recovering information an easier manner thus they can develop tools focused user.