Community detection algorithms of Clauset, Newman, and Moore (CNM) is a method of eliminating the influence of the community defined in each step of the algorithm. CNM algorithm is algorithm in cluster ranks with time running on a network with n vertices and m edges is O (mdlogn) with d is the depth of the dendrogram described the structure of the community. Although the algorithm for time-CNM running fast and quality measure divided the community consistent with the actual model, however the results give off quite a lot of communities have large structures, and the maximization of value modularity can not help us affirm community structure in graphs unless the communities find is the clique.From the weakness of the algorithm CNM then INCRE algorithm COMM EXTRACTION (INC) was developed to reduce the size of the community, the relationships between the objects in the community. In the post will model the network with the interest of the users based on the deductive heterogeneous messages from media activities on social networks such as the user's tweet and comnent on Facebook and Twitter. Level of interest of the user is the interactive content received from the mobile network in nature formed and developed through space and time to do the work around users will create is the social circle.Experimental results show that the algorithm INC. can detect the community is much better than the algorithm CNM. Combination model of user's interest level and the method of extracting the community can determine the community under each topic.
đang được dịch, vui lòng đợi..