Comodularity and detection of co-communities.

This paper introduces the notion of comodularity, to cocluster observations of bipartite networks into co-communities. The task of coclustering is to group together nodes of one type with nodes of another type, according to the interactions that are the most similar. The measure of comodularity is introduced to assess the strength of co-communities, as well as to arrange the representation of nodes and clusters for visualization, and to define an objective function for optimization. We demonstrate the usefulness of our proposed methodology on simulated data, and with examples from genomics and consumer-product reviews.