Structural inference for uncertain networks

摘要: In the study of networked systems such as biological, technological, and social networks available data are often uncertain. Rather than knowing structure a network exactly, we know connections between nodes only with certain probability. this paper develop methods for analysis uncertain data, focusing particularly on problem community detection. We give principled maximum-likelihood method inferring demonstrate how results can be used to make improved estimates true network. Using computer-generated benchmark that our able reconstruct known communities more accurately previous approaches based thresholding. also an example application detection in protein-protein interaction