Coarse graining of complex networks: A k-means clustering approach

Complex networks have been at the forefront of scientific research for more than a decade. A big challenge in complex networks is the share size of the considered systems, especially with the arriving of the era of big data. Coarse graining of complex networks is a possible way to overcome such difficulty. This paper tries to develop a new coarse graining method for complex networks, which is based on the well-known k-means clustering technique. Investigations on some artificial complex networks indicate that the proposed method can significantly reduce the network size and complication, meanwhile, some properties of the considered networks can be preserved to some extent. Moreover, the proposed algorithm allows people to freely choose the sizes of the reduced networks. The associated investigations have potential implications in the analysis and control of large-scale complex networks.