The traditional Bayesian network is relatively fixed, the set of nodes, and intensity dependence relationships are rarely change, thus, it is unable to reflect changes in the actual network state. Such an inaccurate network model is also difficult to inference subsequent network. In order to solve the problem of inference is not accurate enough in traditional model, in this paper, the Markov changes of node parameters with time based on Bayesian network is studied. Next, an adaptive inference

more »

... mpling strategy is put forward and an adaptive inference model based on Bayesian network is designed, then proposes an adaptive Bayesian network Inference algorithm. Finally, simulation results show the effectiveness of the proposed algorithm compared with other algorithms.