When mobile device users meet at a certain place, they may share information obtained from some other places. Since people are more likely to share information if they can benefit from the sharing or if they think the information is of interest to other people, there might exist communities where people share information more often with community members. The communities in mobile social networks represent real social groups where connections are built when people encounter. In this paper, we consider the location and time related to the shared information as the social context in mobile social networks. We propose context-aware community structure that groups people who are more likely to influence each other (i.e., share information with each other in certain contexts) into the same communities. Further, we provide a context-aware community-based user participation strategy in information influence that can reduce unnecessary influence cost. Our evaluation results show that the context-aware community structure is constructed with high internal pairwise similarity, reasonable average community size, and reasonable number of communities. Further, the community-based user participation strategy provides both high average influences and high influence efficiency.