Advances in scaling community discovery methods for signed graph networks

Abstract Community detection is a common task in social network analysis with applications variety of fields including medicine, criminology and business. Despite the popularity community detection, there no clear consensus on most effective methodology for signed networks. In this article, we summarize development networks evaluate current state-of-the-art techniques several real-world datasets. First, give comprehensive background graphs. Next, compare various adaptations Laplacian matrix recovering ground-truth labels via spectral clustering small graph Then, scalability leading algorithms small, large, dense sparse We conclude discussion our novel findings recommendations extensions improvements discovery