Discovering important nodes through comprehensive assessment theory on enron email database

One major problems in the field of social network analysis is how to discover the important and influential nodes based on the network structure. To address this challenge, we propose and use some measures to measure node importance, such as degree measure, improved cluster coefficient measure and a new ranking method based on reputation. Thinking of the unilateral influence of the single measure, we exploit a comprehensive assessment model to synthesize the three measures and discover the interesting and important nodes in the email communication network graph. The experimental results on Enron email dataset show our method is effective and performs better on the problem of important nodes discovery than other measures.