Network community discovery: solving modularity clustering via normalized cut

Modularity clustering is a recently introduced clustering objective function for graph clustering. It has been widely used in bioinformatics and social networks. Its relation to data mining field has yet to be explored. In this paper, we show that a normalized version modularity clustering is identical to the popular normalized cut spectral clustering. This also provides an effective algorithm to solve the modularity clustering problem. We demonstrate this algorithm on several datasets.