A Compressive Sensing Approach to Community Detection with Applications
This provides a more efficient algorithm for graph clustering, beneficial for researchers and practitioners in network analysis, though it is incremental as it builds on existing compressive sensing and thresholding techniques.
The paper tackles the community detection problem in graphs by reformulating it as a sparse linear system and developing a two-stage compressive sensing algorithm, achieving O(n log(n) n_0) operations to find a single cluster and fewer than O(n^2 ln(n)) for all clusters, compared to O(n^3) for spectral clustering.
The community detection problem for graphs asks one to partition the n vertices V of a graph G into k communities, or clusters, such that there are many intracluster edges and few intercluster edges. Of course this is equivalent to finding a permutation matrix P such that, if A denotes the adjacency matrix of G, then PAP^T is approximately block diagonal. As there are k^n possible partitions of n vertices into k subsets, directly determining the optimal clustering is clearly infeasible. Instead one seeks to solve a more tractable approximation to the clustering problem. In this paper we reformulate the community detection problem via sparse solution of a linear system associated with the Laplacian of a graph G and then develop a two-stage approach based on a thresholding technique and a compressive sensing algorithm to find a sparse solution which corresponds to the community containing a vertex of interest in G. Crucially, our approach results in an algorithm which is able to find a single cluster of size n_0 in O(nlog(n)n_0) operations and all k clusters in fewer than O(n^2ln(n)) operations. This is a marked improvement over the classic spectral clustering algorithm, which is unable to find a single cluster at a time and takes approximately O(n^3) operations to find all k clusters. Moreover, we are able to provide robust guarantees of success for the case where G is drawn at random from the Stochastic Block Model, a popular model for graphs with clusters. Extensive numerical results are also provided, showing the efficacy of our algorithm on both synthetic and real-world data sets.