A Graph-Theoretic Framework to Identify Overlapping Communities in Heterogeneous Network
As the advanced data analysis technique, community mining in a complex network has become popular in recent years. However, most of the existing community detection algorithms are based on homogeneous type of graph in which nodes represent individual objects and edges represent a single definition of a relationship, i.e., all edges are of the same type. In reality, an object has the existence of more than one type of relationships; which means multi-relational network. This is a network with a heterogeneous set of edge labels which can represent various relationships.