Automated Discovery of Human Activities inside Pervasive Living Spaces
ABSTRACT The recognition and detection of human activities constitutes a very important step towards the fulfilment of the notion of pervasive environments. By detecting patterns on those behaviours, an environment can adapt and respond to the inhabitants' needs, thus improving the quality of life. This paper presents a framework in which those ideas can be applied and tested. It includes a system using a temporal neural-network driven embedded agent working with online, real-time data from a network of unobtrusive low-level sensors situated in either a simulated environment or a fully fitted real environment such as a whole flat
- SourceAvailable from: Jennifer Boger[show abstract] [hide abstract]
ABSTRACT: Dementia affects an older adult's ability to complete activities of daily living independently. It is envisioned that the principles of pervasive and ubiquitous computing can be applied to develop an intelligent environment that can assist an older adult overcome the problems that they face, and hence i m-prove independence and quality of life at home. This paper presents research being conducted on such a system, and the progress of its development.
Conference Proceeding: Time in Connectionist Models.[show abstract] [hide abstract]
ABSTRACT: IntroductionThe prototypical use of "classical" connectionist models (including the multilayerperceptron (MLP), the Hopfield network and the Kohonen self-organizingmap) concerns static data processing. These classical models are not well suitedto working with data varying over time. In response to this, temporal connectionistmodels have appeared and constitute a continuously growing researchfield. The purpose of this chapter is to present the main aspects of this researcharea and to...Sequence Learning - Paradigms, Algorithms, and Applications; 01/2001
- [show abstract] [hide abstract]
ABSTRACT: This work aims to realise the vision of ambient intelligence in health care environments. The proposed system combines the use of unobtrusive sensors and effectors with intelligent embedded-agents. This paper presents a novel embedded agent mechanism based on an adaptive neural approach which is able to recognize activities inside an environment in an on-line mode. Its ultimate goal is to learn, discriminate and react to personal behaviours and signal departures from the normal behaviour that are significant to health care applications. Experiments were performed in a pervasive computing test-bed known as the iDorm. The presented results show that the system is able to characterize a normal set of activities as well as detecting novel or abnormal activities inside an environment.Systems, Man and Cybernetics, 2004 IEEE International Conference on; 11/2004