Detecting of communities in complex networks with two partitioning approach

Discovery of community structures in complex network is a fundamental task in many fields,for instrance,social science,technology science and biology science.These community structures imply information about system function,and it can be used to help people understand the function of network and its growth mechanism.We optimize modularity density function to spectral questions,and then propose a two partitioning algorithm which uses the leading eigenvectors of the modularity density matrix to split a network into communities.The algorithm is illustrated and compared with spectral clustering based on modularity(Q) using a classic computer generated network.The experimental results show that the proposed approach is effective,particularly when community structure is obscure.