Minimally supervised, iterative, template-based activity learning with body-worn sensors
The aim of this project is the development, implementation, and evaluation of methods for minimally supervised, iterative 'self training' of context recognition systems based on small numbers of sensors in an assisted living application. A target sensor system combining smart phone (equipped with acceleration sensors, microphone, and WLAN interface), home WLAN network with locating capabilities, and ambient sensors for power and water consumption in a flat is foreseen. The system will allow to monitor a variety of individual activities and optimise performance for these specifically, as opposed to full daily routine models. Technical goals are (1) using template based models that map behaviour onto abstract relationships between different signals, (2) iterative clustering and matching of simple situations, and (3) confirmation of the match with an external expert.
Interested candidates are invited to submit their application material for one or more positions within the network. In addition, applicants are asked to provide a 1-page statement on their motivation to become part of this exciting network in PDF format by email to [email protected] The statement may be reviewed by all iCareNet partner organisations. Positions will be filled as soon as possible.
- The researcher can be a national of any country.
- The researcher has not lived, worked or studied in the country of the host organisation of this position for more than 12 months in the 3 years immediately prior to the reference date. Short stays, such as holidays, are not taken into account.
This PhD position is one of the 19 prestigious Marie Curie fellowships for PhD and Postdoc positions that are available within iCareNet.