Towards addressing unauthorized sharing of subscriptions

Zhang, Wei; Challis, Chris

Subscription-based business is booming in recent years, especially in the entertainment sector such as video and music streaming. Usually one subscription account can be shared among family members for the convenience of subscribers. However, account sharing also creates challenges for service provider, as many account owners share their subscriptions outside of the household. The widely spread practice of unauthorized sharing causes huge revenue loss for service providers. However, service providers are very cautious to pursue violators because identifying unauthorized shared accounts is a challenging task. First, the sheer volume of unstructured and noisy data makes it prohibitive to manually process the data. Moreover, it is legitimate for family members to share an account from any location and use many devices as they want. It is tricky to differentiate between unauthorized and legitimate sharing. In this paper, we propose an efficient solution to address the account sharing problem. Based on usage log data, our solution builds user profiles by accumulating and representing geolocation and device usage information. Then we estimate the risk of unauthorized sharing by analyzing the usage pattern of each account. The proposed solution can identify a large number of shared accounts and help service providers to recoup a significant amount of lost revenue.