Zhou M.,Chongqing University |
Zhou M.,Key Laboratory of Software Engineering of Chongqing |
Zhu Q.,Chongqing University |
Zhu Q.,Key Laboratory of Software Engineering of Chongqing |
And 4 more authors.
Journal of Information and Computational Science | Year: 2010
Real networks have generally a nature of community structure: the connection between nodes in each community is close but the connection between nodes in different communities is sparse. In this paper the Semantic Gossip is used to describle the preference of node. Than a Gossip Interest Attenuation Mechanism and a Network Evolution Model are proposed. In the model, a preference is firstly spread to each node in the network, and the receiving node carries on the semantic matching. Then the connectivity between nodes and the network topology are adjusted by the preference similarity. Finally the optimized community network structure is explored so as to set the super nodes. Every node makes decisions based on its preference, which reduce the dependence on the global knowledge. Copyright © 2010 Binary Information Press.