Activity location inference of users based on social relationship

Users in social networks often form relationships with other users who participate together in various activities nearby. The activity locations which are frequently shared with the friends are important in real life in order to understand the precise spatial space of the social users. However, the locations of individuals in a social network are often unknown. This is because the social users do not bother to broadcast their locations in public due to many reasons including privacy. Identifying the top activity location of a user at a higher granularity level will improve various community based applications like Meetup, Groupon, etc. In this paper, we propose a method to infer the top activity location of social users using the implicit information available in the network. Our proposed approach can estimate the activity location of a user by propagating the spatial information of the neighbors through friendship edges. We maintain a proper inference sequence to propagate the location labels of the users. We find that the proposed method has significantly improved the state-of-the-art network based location inference techniques in terms of both the accuracy and efficiency.

Automated summary

Users in social networks form relationships and share activity locations with friends, but individual locations are often unknown due to privacy concerns. This paper proposes a method to infer top activity locations using implicit information in the network, resulting in significantly improved accuracy and efficiency compared to existing techniques.