Efficient Content Distribution in DOOH Advertising Networks Exploiting Urban Geo-Social Connectivity

Digital out-of-home (DOOH) advertising networks, comprised of pervasive distributed digital signages (screens), are rapidly growing. It is reported that more than 70% of DOOH revenue comes from local ads, while it is especially challenging to decide when and where to deliver the most suitable ad due to the spatio-temporal dynamics of human mobility and preferences. Understanding urban geo-social connectivity in terms of people movement would greatly benefit ad content distribution, and could potentially be utilized by a large number of mobile applications and geo-social services. However, existing DOOH ad distribution systems are designed to target individuals, which might not be the best choice in public spaces, and do not consider the preferences of "cohort of users". In this paper, we propose an alternative approach to target cohort of users extracting urban geo-social connectivity through large-scale mobile network data and existing geo-social service data. We construct a dynamic urban geo-social connectivity graph, and formulate the problem of distributing ads for maximum exposure to the "right" users under a constrained budget. Hence, we propose a heuristic algorithm. Simulation results show that our system targeting "cohort of users" achieves a maximum 300% improvement compared to naive distributing method in displaying ads to the "right people" when user preferences are completely known, while a minimum of 25% improvement when the knowledge of user preferences is limited.