MURAL - Maynooth University Research Archive Library

    Habitualisation: localisation without location data

    McGrath, Rory and Coffey, Cathal and Pozdnoukhov, Alexei (2012) Habitualisation: localisation without location data. In: Nokia MDC challenge at PERVASIVE'2012, June 18-22 2012, Newcastle University, UK. (Unpublished)

    [img] Download (1MB)

    Share your research

    Twitter Facebook LinkedIn GooglePlus Email more...

    Add this article to your Mendeley library


    This paper looks at identifying the locations of users from the Nokia MDC dataset throughout the day without taking into consideration location based data. By looking at a users habits and idiosyncrasies we determined the likelihood of a users location within known stay regions which we call habitats. The features used to determine location were extracted from a users interaction with the smart phone. None of the features contained a users locations or a users proximity to objects with known locations. Using a set of structured output support vector learning techniques we found that a users location with respect to the areas of typical activities is well predictable solely from daily routines and a smart phone usage habits.

    Item Type: Conference or Workshop Item (Paper)
    Additional Information: Research presented in this paper was funded in part by Science Foundation Ireland Strategic Research Cluster grant 07/SRC/I1168 and 11/RFP.1/CMS/3247 award, and IBM PhD Fellowship program. The authors gratefully thank Aonghus Lawlor and Felix Kling for their support, fruitful discussions and help with software.
    Keywords: machine learning; kernel methods; smart cities; pervasive computing;
    Academic Unit: Faculty of Science and Engineering > Research Institutes > National Centre for Geocomputation, NCG
    Item ID: 3928
    Depositing User: Dr Alexei Pozdnoukhov
    Date Deposited: 04 Oct 2012 08:50
    Refereed: No
    Funders: Science Foundation Ireland, IBM

    Repository Staff Only(login required)

    View Item Item control page


    Downloads per month over past year