Research - Dr. Susan McKeever

separator bar

Office based data set

The following dataset was collected in the Computer and Adaptive Systems Lab in UCD. 

A user is monitored for a week in the office, monitored by three sensors. Full details and download of the data set are available from here

separator bar

PhD (2011)

Situation Recognition (in pervasive systems) using extended Dempster Shafer theory  PDF

This  work focuses on recognising people's activities in smart environments using sensors.  With sensors embedded in everyday objects, people's activities can be automatically  and remotely tracked.  This will be useful, for example, in allowing elderly residents to live independently, but with the support of a monitoring system in their home  As part of such a system, data from sensors has to be interpreted and recognised as higher level situations, such as whether a person is eating or resting and so on.  This is a difficult problem because of the unreliability of sensors, changing habits of people and general uncertainty of the recognition process. My work developed a reasoning technique, based on evidence (Dempster Shafer) theory.  This approach is ideal for using sensor data, but does not rely on training data, unlike the machine learning approaches that are commonly used to address this problem.  As part of this work, I extended existing evidence theory to allow time and sensor quality knowledge into the reasoning process. I prove that the approach is effective for situation recognition, and in particular, that my  extensions to evidence theory significantly improve recognition.

separator bar

Masters in IT (2003)

A strategic decision tool for Business to Business e-commerce  PDF

separator bar

Publications

Ye. J, Dobson S., McKeever S.,(2010), Situation Identification Techniques in pervasive computing: a review, Journal of  Pervasive and Mobile Computing PDF

McKeever S, Ye, J, Coyle L, Bleakley C, Dobson, S (2010), Activity Recognition using Temporal Evidence theory, Journal of Ambient Intelligence and Smart Environments, July 2010 To appear. PDF

Juan Ye, Lorcan Coyle, Susan McKeever, and Simon Dobson (2010).  Dealing with activities with diffuse boundaries. Proceedings of Pervasive
2010 workshop on How to do good activity recognition research? Experimental methodologies, evaluation metrics, and reproducibility issues .
Helsinki, Finland. May 17-21, 2010. PDF

McKeever S, Ye, J, Coyle L, Dobson, S (2009) Using Dempster Shafer Theory of Evidence for Situation Inference, Proceedings of EuroSSC 2009, Sept 2009, London, UK.PDF

McKeever S, Ye, J, Coyle L, Dobson, S (2009) A Context Quality Model to Support Transparent Reasoning with Uncertain Context, Proceedings of QuaCon 2009, June 2009, Stuttgart, Germany PDF

Lorcan Coyle; Juan Ye; Susan McKeever; Stephen Knox; Matthew Stabeler; Simon Dobson; Paddy Nixon. Gathering Datasets for Activity IdentificationDeveloping. Shared Home Behavior Datasets to Advance HCI and Ubiquitous Computing Research. Workshop at CHI 2009, Boston, USA (In Press)  PDF

Juan Ye, Susan McKeever, Lorcan Coyle, Steve Neely, and Simon Dobson (2008). Resolving Uncertainty in Context Integration and Abstraction. ICPS 2008: Proceedings of the International Conference on Pervasive Services, ACM, June 2008. To Appear. PDF

McKeever S, Ye, J, Coyle L, Dobson, S (2008) A Multilayered Uncertainty Model for Context Aware Systems, Pervasive 2008 Late Breaking Results, May 2008, Sydney, Australia PDF

McKeever, S. Defining the Range of B2B E-Commerce Formats. Proceedings of the IADIS International Conference WWW/Internet 2003, ICWI 2003,

McKeever, S. Understanding Web Context Management Systems: Evolution, lifecycle and market. Journal of Industrial Management and Data Systems 103(9): 686-692