Community Detection Using Maximum Connection Probability in Opportunistic Network

A novel approach is proposed in this paper to detect community structure in opportunistic networks. Different from the existing solutions, this approach uses Maximum Connection Probability (MCP) instead of encounter probability. This approach is established in two phases. Firstly, an algorithm is proposed to derive the MCP of any node to other nodes. Secondly, the community structure derived from the MCP is identified using a divisive algorithm. Simulation is conducted based on walking day movement model to evaluate the approach. The results show that the proposed approach can detect community structure more accurately and reflect human relationship in reality.