Mobile Computing, Applications, and Services. 4th International Conference, MobiCASE 2012, Seattle, WA, USA, October 11-12, 2012. Revised Selected Papers

Research Article

Achieving Targeted Mobile Advertisements While Respecting Privacy

Download
432 downloads
  • @INPROCEEDINGS{10.1007/978-3-642-36632-1_14,
        author={Elia Palme and Basil Hess and Juliana Sutanto},
        title={Achieving Targeted Mobile Advertisements While Respecting Privacy},
        proceedings={Mobile Computing, Applications, and Services. 4th International Conference, MobiCASE 2012, Seattle, WA, USA, October 11-12, 2012. Revised Selected Papers},
        proceedings_a={MOBICASE},
        year={2013},
        month={2},
        keywords={mobile pervasive advertisement privacy targeted one-to-one marketing mobile agent},
        doi={10.1007/978-3-642-36632-1_14}
    }
    
  • Elia Palme
    Basil Hess
    Juliana Sutanto
    Year: 2013
    Achieving Targeted Mobile Advertisements While Respecting Privacy
    MOBICASE
    Springer
    DOI: 10.1007/978-3-642-36632-1_14
Elia Palme,*, Basil Hess1,*, Juliana Sutanto1,*
  • 1: ETH Zürich
*Contact email: epalme@ethz.ch, bhess@ethz.ch, jsutanto@ethz.ch

Abstract

Broadcasted mobile advertisements are increasingly being replaced by targeted mobile advertisements through consumer profiling. However privacy is a growing concern among consumers who may eventually prevent the advertising companies from profiling them. This paper proposes an agent-based targeting algorithm that is able to guarantee full consumer privacy while achieving mobile targeted advertising. We implemented a grocery discount-discovery application for iPhone that makes use of the new approach. We show that on modern hardware like on the iPhone, it’s feasible to run a client-based and privacy-preserving targeting algorithm with minimal additional computational overhead compared to a random advertising approach. We evaluated the targeting method by conducting a large-scale field-experiment with 903 participants. Results show that the computational overhead on user devices is well tolerated, compared to the control group with randomized advertising the targeting group showed a significantly increased application usage of 18%.