Modeling the value of information granularity in targeted advertising

Swati Rallapalli, Qiang Ma, Han Hee Song, Mario Baldi, S. Muthukrishnan, Lili Qiu

    Research output: Contribution to journalArticlepeer-review

    Abstract

    Behavioral Targeting (BT) in the past few years has seen a great upsurge in commercial as well as research interest. To make advertising campaigns more effective, advertisers look to target more relevant users. Ad-networks and other data collectors, such as, Cellular Service Providers (CSPs), hold a treasure trove of user information that is extremely valuable to advertisers. Moreover, these players may have complimentary sets of data. Combining and using data from different collectors can be very useful for advertising. However, in the trade of data among the various players, it is currently unclear how a price can be attached to a certain piece of information. This work contributes (i) a MOdel of the Value of INformation Granularity (MoVInG) that captures the impact of additional information on the revenue from targeted ads in case of uniform bidding and (ii) an expression that is applicable in more general scenarios. We apply MoVInG to a user data-set from a large CSP to evaluate the financial benefit of precise user data.

    Original languageEnglish (US)
    Pages (from-to)42-45
    Number of pages4
    JournalPerformance Evaluation Review
    Volume41
    Issue number4
    DOIs
    StatePublished - Mar 2014

    ASJC Scopus subject areas

    • Software
    • Hardware and Architecture
    • Computer Networks and Communications

    Fingerprint Dive into the research topics of 'Modeling the value of information granularity in targeted advertising'. Together they form a unique fingerprint.

    Cite this