Hierarchical, Parameter-Free Community Discovery

Abstract

Abstract. Given a large bipartite graph (like document-term, or userproduct graph), how can we find meaningful communities, quickly, and automatically? We propose to look for community hierarchies, with communities-within-communities. Our proposed method, the Context-specific Cluster Tree (CCT) finds such communities at multiple levels, with no user intervention, based on information theoretic principles (MDL). More specifically, it partitions the graph into progressively more refined subgraphs, allowing users to quickly navigate from the global, coarse structure of a graph to more focused and local patterns. As a fringe benefit, and also as an additional indication of its quality, it also achieves better compression than typical, non-hierarchical methods. We demonstrate its scalability and effectiveness on real, large graphs.

D-cores: Measuring Collaboration of Directed Graphs Based on Degeneracy

“…For Wikipedia 2004, the balanced collaboration index(BCI) value is 38, while the respective D-core DC 38,38 contains 237 nodes. For the same digraph, the inherent collaboration index ICI is 36 and is obtained for the D-cores DC 39,33 that contains 206 nodes. For the OCI index, we obtain two OCIoptimal frontier cells corresponding to the DC 38,41 and DC 36,43 D-cores containing 228 and 233 nodes, respectively.…”

“…Such are "Betweenness" [43], "Centrality" [39], Clustering coefficient (a measure of the likelihood that two associates of a node are associates themselves).…”

Self-organizing kernel adaptive filtering

“…Starting from an overly complex model and then pruning unneeded basis function according to MDL, Leonardis and Bischof [39] proposed a radial basis function (RBF) network formulation to balance accuracy performance, training time, and network complexity. Besides neural networks, MDL has also been successfully used in vector quantization [40], clustering [31,41,42], graphs [43,44], and so on. In most cases, MDL is used for supervised learning as a penalty term on the error function or as a criterion for model selection [40].…”

Enabling community-driven information integration through clustering

“…In a social network setting, the graph topology provides information about the relationships between individuals, whereas vertex properties describe the role of individuals. Papadimitriou et al [26] also show that clustering can be used to identify a hierarchy of user communities in large graphs. Specifically, they developed an algorithm that constructs recursive community structures.…”