Community Structure in Multi-Mode Networks: Applying an Eigenspectrum Approach*

We combine the logic of multi-mode networks developed in Fararo and Doreian (1984) with Newman’s (2006) spectral partitioning of graphs into communities. The resulting generalization of spectral partitioning provides a simple, elegant, and useful tool for discovering the community structure of multi-mode graphs. We apply the generalized procedure to a published threemode network and find that the results of the algorithm are consistent with existing substantive knowledge. We also report the results of extensive simulations, which reveal that the generalization becomes more effective as the networks become denser. Authors David Melamed is currently an Assistant Professor of Sociology at the University of South Carolina. His research interests are in group processes, social networks, mathematical modeling, and stratification. Ronald L. Breiger is Professor of Sociology at the University of Arizona and holds affiliate appointments in the Graduate Interdisciplinary Program in Statistics and in the School of Government and Public Policy. His research interests include social network modeling, culture, and adversarial networks. A. Joseph West is a PhD Candidate in Sociology at the University of Arizona. His research interests lie at the intersection of social networks, culture, and social movements. Please address correspondence to David Melamed, Department of Sociology, University of South Carolina, Sloan College 321, 911 Pickens Street, Columbia, SC 29208. Email: [email protected] Ronald L. Breiger University of Arizona A. Joseph West University of Arizona Notes This research was supported by grants from the Defense Threat Reduction Agency (HDTRA1-10-1-0017) and the Air Force Office of Scientific Research (FA9550-10-1-0569) for which the second-listed author is the PI and a Co-PI, respectively. We thank Sean Everton and two anonymous reviewers for helpful feedback on this project.