The Study Of Community Detecting On Directed And Weighted Email Network

SNA(Social Network Analysis) is a new hot spot in data mining area. It is a combination of individuals, or organizations and the connections between them. Through the study of social network theory, we can try to discover the relationships hidden in these communications, and apply it to e-commerce, information recommendation and so on.With the development of Internet, email has become an important and popular mean of communication. Email, as an important part of social network, has become the powerful communication platform for cooperation and knowledge sharing. Email mining is an emerging technique in the area of data mining. Its appearance stems from the broad use of email and the daily mature of data mining technique. However, how to figure out the communities as well as their core-persons from a great deal of emails is a hard problem.In this paper, we propose a solution to discover the specific communities of email network, analyze the feature of broadcasting email and the relationship between email sender and receiver which based on weighted and directed email community. Then analyze the structure of weighted and directed email network, to determining the community structure of broadcasting. To finding the overlapping communities is another advantage of this algorithm.Next, after the analysis of the above broadcasting community, we propose a detecting method of organizational structure which is based on important nodes. We introduce the HITS algorithm and propose the computational method of leader value based on the improved HITS algorithm. Then by detecting the leader-member relation in the email network, construct the chart of organizational structure.Finally, the experimental results on Enron email dataset show our method is effective on the problem of detecting the specific structural type of community in the email network.