Potential collaboration discovery using document clustering and community structure detection

Complex network analysis is a growing research area in a wide variety of domains and has recently become closely associated with data, text and web mining. One of the most active areas in the study of complex networks is the detection of community structure, which can be related to the clustering problem in data mining. This paper employs a community structure detection algorithm for document clustering in order to discover potential relationships in a social network. The proposed approach is explored in a case study of potential collaboration discovery among the research staff in the Graduate Civil Engineering Department of the Federal University of Rio de Janeiro, Brazil. The results show that the combined use of both techniques provides useful insights on the relationships, both existent and potential, among individuals in the social network.