A Literature Survey on Label Propagation for Community Detection
Laassem Brahim, Loubna Boujlaleb, Ali Idarrou · 2021
Community detection (CD) aims to discover cohesive groups or clusters in a network, it’s a fundamental task in social network analysis. Label propagation (LPA) is a semi-supervised machine learning algorithm where labels are assigned to unlabeled nodes by propagating labels through the graph representation of a network to uncover communities. This article presents an organized and comprehensive overview of important algorithms and approaches to the community detection problem based on LPA. We discuss the basic concepts of such an approach and a toy example is given to illustrate the process of LPA. Furthermore, a brief overview of benchmark datasets and evaluation metrics used in this field is also provided. Finally, we highlight current research directions and open problems.