Abstract: This paper focuses on finding the same andsimilar users based on location-visitation data in a mobile environment. Wepropose a new design that uses consumer-location data from mobile devices(smartphones, smart pads, laptops, etc.) to build a "geosimilaritynetwork" among users. The geosimilarity network (GSN) could be used for avariety of analytics-driven applications, such as targeting advertisements tothe same user on different devices or to users with similar tastes, and toimprove online interactions by selecting users with similar tastes. The basicidea is that two devices are similar, and thereby connected in the GSN, whenthey share at least one visited location. They are more similar as they visitmore shared locations and as the locations they share are visited by fewerpeople. This paper first introduces the main ideas and ties them to theory andrelated work. It next introduces a specific design for selecting entities withsimilar location distributions, the results of which are shown using realmobile location data across seven ad exchanges. We focus on two high-levelquestions: (1) Does geosimilarity allow us to find different entitiescorresponding to the same individual, for example, as seen through differentbidding systems? And (2) do entities linked by similarities in local mobilebehavior show similar interests, as measured by visits to particularpublishers? The results show positive results for both. Specifically, for (1),even with the data sample's limited observability, 70%-80% of the time the sameindividual is connected to herself in the GSN. For (2), the GSN neighbors ofvisitors to a wide variety of publishers are substantially more likely also tovisit those same publishers. Highly similar GSN neighbors show very substantiallift.
