A Unified Community Detection, Visualization and Analysis Method

Abstract

With the widespread of social networks on the Internet, community detection in social graphs has recently become an important research domain. Interest was initially limited to unipartite graph inputs and partitioned community outputs. More recently bipartite graphs, directed graphs and overlapping communities have all been investigated. Few contributions however have encompassed all three types of graphs simultaneously. In this paper we present a method that unifies community detection for these three types of graphs while at the same time merges partitioned and overlapping communities. Moreover, the results are visualized in a way that allows for analysis and semantic interpretation. For validation purposes this method is experimented on some well-known simple benchmarks and then applied to real data: photos and tags in Facebook and Human Brain Tractography data. This last application leads to the possibility of applying community detection methods to other fields such as data analysis with original enhanced performances.

Efficient community detection of network flows for varying Markov times and bipartite networks

“…For undirected networks, the flow is equal in both directions such that the bipartite leap dynamics correspond to Markov time t = 1 in Eqs. (9) and (10). That is, the bipartite leap dynamics effectively correspond to the standard unipartite dynamics in which the node type is ignored as shown in Fig.…”

Overlapping Community Detection Optimization and Nash Equilibrium

“…Other original methods use well-known results in Galois Lattices ( [6] and [32]), though their algorithms and representations remain complex. In a recent article [7], we demonstrated the possibility of unifying unipartite, bipartite and oriented graphs, in addition to extracting both overlapping and partitioned communities; the overlapping properties are computed using a simple simultaneous membership function. We decided to apply the well-known Louvain Algorithm [4], which iteratively aggregates the graph vertices in order to maximize the modularity function.…”

“…We have shown in [7] that detecting partitioned communities in unipartite, bipartite or directed graphs may be considered as community detection in unipartite graphs. Consequently, we introduced definitions and proved all properties on unipartite graphs, before applying them to the unipartite and bipartite graphs in our experiments.…”

“…Fortunato [10] however demonstrated that this function tends to merge small communities and, consequently, swallows small granularities. For bipartite graphs (and directed graphs, which can be reduced to bipartite graphs, see [7]), other formulations have been offered. In particular, Barber's formulation [2] has won consensus among researchers: it is not very different from Newman's.…”

“…In a previous article we focused on extracting overlapping communities in bipartite graphs through the use of Galois Lattices [6]. In a more recent paper, we proposed a different method [7] that uses any partitioning algorithm for unipartite graphs in order to process any type of graph (i.e. Unipartite, Bipartite, Oriented) and then extract partitioned and overlapping communities.…”