9th IEEE International Conference on Collaborative Computing: Networking, Applications and Worksharing

Research Article

Answering Complex Location-Based Queries with Crowdsourcing

Download664 downloads
  • @INPROCEEDINGS{10.4108/icst.collaboratecom.2013.254104,
        author={Karim Benouaret and Raman Valliyur-Ramalingam and Francois Charoy},
        title={Answering Complex  Location-Based Queries with Crowdsourcing},
        proceedings={9th IEEE International Conference on Collaborative Computing: Networking, Applications and Worksharing},
        publisher={ICST},
        proceedings_a={COLLABORATECOM},
        year={2013},
        month={11},
        keywords={crowdsourcing location-based queries query transformation process management},
        doi={10.4108/icst.collaboratecom.2013.254104}
    }
    
  • Karim Benouaret
    Raman Valliyur-Ramalingam
    Francois Charoy
    Year: 2013
    Answering Complex Location-Based Queries with Crowdsourcing
    COLLABORATECOM
    IEEE
    DOI: 10.4108/icst.collaboratecom.2013.254104
Karim Benouaret1, Raman Valliyur-Ramalingam2, Francois Charoy2,*
  • 1: Inria
  • 2: LORIA - Inria - University of Lorraine - CNRS
*Contact email: francois.charoy@loria.fr

Abstract

Crowdsourcing platforms provide powerful means to execute queries that require some human knowledge, intelligence and experience instead of just automated machine computation, such as image recognition, data filtering and labeling. With the development of mobile devices and the rapid prevalence of smartphones that boosted mobile Internet access, location-based crowdsourcing is quickly becoming ubiquitous, enabling location-based queries assigned to and performed by humans. In sharp contrast of existing location-based crowdsourcing approaches that focus on simple queries, in this paper, we describe a crowdsourcing process model that supports queries including several crowd activities, and can be applied in a variety of location-based crowdsourcing scenarios. We also propose different strategies for managing this crowdsourcing process. Finally, we describe the architecture of our system, and present an experimental study conducted on pseudo-real dataset that evaluates the process outcomes depending on these execution strategies.