Partitioning multimode networks

Most networks examined so far involve connections between nodes all of the same type, known as one-mode networks. This chapter examines partitioning and clustering in multimode network data. A number of techniques have been developed for dealing with non-binary data or more precisely non-network type data. The chapter first concentrates on two mode datasets and then discusses general multimode approaches. In considering two-mode networks the authors consider the problem of partitioning both modes to find sets of actors and events. For single-mode networks the most commonly used and accepted technique is Newman’s community detection, which optimizes modularity. M. J. Barber extended modularity to two-mode data and developed an algorithm specifically for this type of data.