Detection Of Normal and Novel Behaviours In Ubiquitous Domestic Environments

• Published in: Computer journal Vol. 53; no. 2; pp. 142 - 151
• Main Authors: Rivera-illingworth, F., Callaghan, V., Hagras, H.
• Format: Journal Article
• Language: English
• Published: Oxford Oxford University Press 01-02-2010; Oxford Publishing Limited (England)
• Subjects: Behavior; Elder care; Human-computer interaction; Neural networks; Quality of life; Sensors; ubiquitous computing; intelligent environments; activities of daily living; neural networks; activity recognition
• ISSN: 0010-4620; 1460-2067
• DOI: 10.1093/comjnl/bxm078

The importance of ubiquitous environments has increased in recent years as it has been recognized as a paradigm that can improve the quality of life of many sectors of the population especially care of elderly people by providing automated environments that adapt and respond to its inhabitants' needs. The aim of the work presented here is to provide a solution to the problem of recognition and detection of human behaviours inside ubiquitous environments by using a neural-network driven embedded agent working with online, real-time data from a network of unobtrusive low-level sensors. The final objective of this system was to classify a 'normal' pattern of activities, and sense deviations from it, which could be employed for home care applications.