Abstract: The proliferation of mobile devices has witnessed the extensive research on opportunistic network in both academia and industry. Community-based opportunistic network combine the social people and wireless communication devices so that characteristics from both the opportunistic network and social networking are inherited. However, the prevalent multi-copy message routing algorithms in opportunistic network will lead to long transmission delay and high message loss rate when deployed directly in the community-based opportunistic network. To tackle the above problems, a community-based message transmission scheme (CMTS) is proposed in utilizing the clustering characteristics of all the nodes in community-based opportunistic network. CMTS utilizes Newman’s weighted network analysis method to organize network nodes into different communities based on the contact frequencies among them. In dealing with intra-community communication, CMTS determines the number of message copies in the network and chooses more active nodes as the relays. For inter-community communication, messages are transmitted to target communities mainly relying on the nodes which have larger social degrees. Simulation results show that CMTS can balance well the tradeoff between delivery ratio and resource consumption in community-based opportunistic network. The evaluation results also show that CMTS can reduce the message forwarding times by about 20% when compared with Spray and Focus.