Abstract: This paper analyzes the existing social network''s history,characteristics and applications,next we derives the corresponding theoretical model．In our fuzzy cluster belief propagation theory,according to social network’s coverage,we classify it into hierarchical sets,analysis the affected number of users and user behaviors,then dividing the sets into different clusters through the proposed algorithm．Since we adopt the fuzzy set to calculate the approximation of propagation probability,the computation complexity is reduced．Then,we give an example of social network for electrical automobile charge,in which the wireless sensor networks and social Web2.0application mode are both used．Through comparing the statistic data and those estimated from our cluster algorithm,it proves that the effective and accurate of algorithm．Finally,we give the general social network business model and look forward to its impact on the future of the Internet.

Key words: Social network,User behavior,Relationship measurement,Fuzzy cluster,Business mode,Sensor network