Fig. 2From: GRANDPA: GeneRAtive network sampling using degree and property augmentation applied to the analysis of partially confidential healthcare networksThe original and generated graphs from Zachary’s Karate network (Zachary 1977). A The original graph, where vertex color indicates a category associated with degree, vertex size represents a category associated with centrality, and community areas are drawn. B The PGM-generated graph, if just categorical attributes and degree augmentation are used to generate the network. C The GRANDPA-generated graph, where attributes, community labels, and degree are used to generate the network. In Fig. 2C, we recaptured similar vertex attribute relationship and nearly identical communities in both overall number of communities, size of communities, and internal community structureBack to article page