A0510
Title: Estimation of the number of communities for sparse networks
Authors: Shirshendu Chatterjee - City University of New York (United States) [presenting]
Sharmodeep Bhattacharyya - Oregon State University (United States)
Neil Hwang - City University of New York (United States)
Jiarui Xu - Meta (United States)
Abstract: Among the nonparametric methods of estimating the number of communities (K) in a community detection problem, methods based on the spectrum of the Bethe Hessian matrices have garnered much popularity for their simplicity, computational efficiency, and robustness to the sparsity of data. For certain heuristic choices of these matrices, such methods were shown to be consistent for semi-dense networks. A consistent K-estimation procedure is developed based on spectral properties of suitable Bethe Hessian matrices in the sparse regime. The performance of the resulting estimation procedure is evaluated theoretically and empirically through extensive simulation studies and application to a comprehensive collection of real-world network data.
