A1274
Title: New directions in community detection for core-periphery networks
Authors: Sinyoung Park - University of Bath (United Kingdom) [presenting]
Sandipan Roy - University of Bath (United Kingdom)
Matthew Nunes - University of Bath (United Kingdom)
Abstract: Spectral clustering has been widely used as a popular tool for community detection in data with a network structure. However, spectral clustering does not perform well on certain network structures, particularly core-periphery networks. To improve clustering performance in core-periphery structures, adjacency spectral embedding (ASE) has been introduced. Despite its advantages, ASE has several limitations including its optimal performance only on dense networks. To address these limitations, a new approach is proposed, called doubled adjacency spectral embedding (DASE). It is demonstrated that DASE overcomes these challenges, particularly highlighting the improved clustering performance on both sparse and dense networks in the presence of core-periphery structures.
