User-Rank: Generic Query Optimization for Participatory Social Applications

Online social networks have become a key communication medium for millions of Internet users. Building on this success a new class of social applications have emerged that use online social networks as a platform to access an established community of users. While some social applications act as a forum for users to interact, an emerging class of participatory social applications creates value by using data received from users to deliver a service. The success of participatory social applications is dependent on user responsiveness to application queries. In this paper we propose ´User-Rank´, an application-independent approach to optimizing the querying of social network users. The design of User-Rank is based upon a 30-day experiment, which involved sending 3,055 messages to 70 users to determine the factors influencing response time and message loss. User-Rank achieves significant performance improvements relative to random querying and is lightweight with minimal memory and computational requirements.