Trajcevski Delivers Keynote Talk at 4th International ACM SIGSPATIAL workshop on Mobile Geographic Information Systems (MobiGIS)

His Keynote Talk is titled, "Fusion Uncertain Location Data from Heterogeneous Sources."

4th International ACM SIGSPATIAL workshop on Mobile Geographic Information Systems

Prof. Goce Trajcevski gave a Keynote Talk, titled, "Fusion Uncertain Location Data from Heterogeneous Sources" at the 4th International ACM SIGSPATIAL workshop on Mobile Geographic Information Systems (MobiGIS), held on Tuesday, November 3 in Seattle, WA.

Prof. Goce TrajcevskiTalk Abstract: The keynote addressed techniques which, when combining different uncertain data may yield tighter boundaries of the whereabouts of moving objects. Combining the functionality of mobile devices (smartphones and tablets), wireless communication (Wi-Fi, Bluetooth and 3/4G), and positioning technologies (GPS, Assisted GPS and GLONASS) results in a new era of mobile geographic information systems (MobiGIS) that aim at providing various invaluable services, including location-based services, intelligent transportation systems, logistics management, security and safety, etc. Many of those services, in turn, rely on efficient techniques for managing the data capturing the information pertaining to the whereabouts-in-time of the moving entities – storing, retrieving and querying such data.

Regardless of how location data is obtained, there is the inevitable component of uncertainty due to (end-points of spectrum): 1. Imperfection of the localization devices (e.g., even differential GPS has errors, and is not applicable for indoor-tracking); and 2. The measurements performed in discrete time-instants are being used to model a continuous motion – entailing the use of some interpolation.

The keynote discussed the potential benefits of combining (i.e.., fusing) location data from heterogeneous sources in various scenarios from geo-social networks and symbolic trajectories, and addressed some alternative measures of the benefits of the fusion of such data in terms of similarity-detection. It also discussed possible ways of improving the existing privacy-related approaches from becoming weaker in the presence of multiple data sources as well as dealing with uncertainty when attempting to extract sentiment from multiple mobile data-feeds.

ACM SIGSPATIAL MobiGIS 2015 brings together researchers and practitioners from the GIS community, the mobile computing community, and the data management community. Many current research areas, such as spatio-temporal databases, spatio-temporal data mining, mobile cloud computing, remote sensing, participatory sensing, or social networks, raise research problems that lie at the boundary between these three communities. MobiGIS's goal is to foster an opportunity for researchers from these three communities to gather and discuss ideas that will shape and influence these emerging GIS-related research areas.