With the growing number of available social and biological networks, the problem of detecting network community structure
is becoming more and more important which acts as the first step to analyze these data. In this paper, we transform network
data so that each node is represented by a vector, our method can handle directed and weighted networks. it also can detect
networks which contain communities with different sizes and degree sequences. This paper reveals that network community can
be formulated as a cluster problem.
This work is supported by the NNSF (10531070) of China and Science Fund for Creative Research Group. The authors thank Dr.
Martin Rosvall in Washington University for giving valuable comments on this paper.