Finding Mobile Consumers with a Privacy-Friendly Geo-Similarity Network

  • David Martens
  • Alan Murray
  • Foster Provost

This paper focuses on finding the same and similar users based on location-visitation data in a mobile environment.  We propose a new design that uses consumer location data from mobile devices (smart phones, smart pads, laptops, etc.) to build a “geo-similarity network” among users.  The geo-similarity network (GSN) could be used for a variety of analytics-driven applications, such as targeting advertisements to the same user on different devices or to users with similar tastes, and improving online interactions by selecting users with similar tastes.  The basic idea is that two devices are similar, and thereby connected in the geo-similarity network, when they share at least one visited location.  They are more similar as they visit more shared locations, and as the locations they share are visited by fewer people.  This paper first introduces the main ideas and ties them to theory and related work.  Next we introduce a specific design for selecting entities with similar location distributions, the results of which are shown using real mobile location data across seven ad exchanges.  We focus on two high-level questions:  (1) does geo-similarity allow us to find different entities corresponding to the same individual, for example as seen through different bidding systems?  And (2) do entities linked by similarities in local mobile behavior show similar interests, as measured by visits to particular publishers?  The results show positive results for both. Specifically, for (1) even with the data sample’s limited observability, 70-80% of the time the same individual is connected to herself in the GSN.  For (2), the GSN neighbors of visitors to a wide variety of publishers are substantially more likely also to visit those same publishers.  Highly similar GSN neighbors show very substantial lift.