Skip to Main content Skip to Navigation
Journal articles

An Intelligent Knowledge System for Designing, Modeling, and Recognizing the Behavior of Elderly People in Smart Space

Abstract : In this paper, a context-sensitive descriptive language is proposed to design and model the daily living activities of elderly people. The objective is to simplify and represent correctly the knowledge collected by sensors (low level) and to have a relevant recognition of the person's knowledge (high level). The proposed language is based on several rules and constraints through intelligent meaning. It is dedicated to a better understanding and semantic design and description of the behavior of elderly people. Subsequently, in order to provide a powerful knowledge recognition system, a hybrid Markov model is proposed to recognize and predict the activities designed by the proposed language. The proposed model is adapted to the reasoning of the new language. This allows providing a hierarchical and temporal relationship within the knowledge. It is responsible to recognize and predict the behavior of the elderly people e ciently. The flexibility and the intelligibility of the proposed language is proven and the accuracy of the recognition model is demonstrated which ensures the e ciency of the proposed knowledge recognition system.
Complete list of metadatas

Cited literature [53 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02967216
Contributor : Tayeb Lemlouma <>
Submitted on : Wednesday, October 14, 2020 - 5:31:34 PM
Last modification on : Tuesday, October 20, 2020 - 6:48:06 PM

File

AIHC_Springer_Journal_Final_HA...
Files produced by the author(s)

Identifiers

Citation

Zaineb Liouane, Tayeb Lemlouma, Philippe Roose, Frédéric Weis, Hassani Messaoud. An Intelligent Knowledge System for Designing, Modeling, and Recognizing the Behavior of Elderly People in Smart Space. Journal of Ambient Intelligence and Humanized Computing, Springer, 2020, ⟨10.1007/s12652-020-01876-5⟩. ⟨hal-02967216⟩

Share

Metrics

Record views

49

Files downloads

8