Bayesian Networks Inference algorithm based on improved Junction Tree Algorithm

The Junction Tree Algorithm (JTA) algorithm is an important inference algorithm in Bayesian Networks (BN). It has a wide range of applications in electromagnetic interference and threat estimation. In order to reduce the complexity of the JTA, this paper proposes Triangulation Optimization Algorithm based on Grey Wolf Optimizer (TOA-GWO) to find the optimal elimination order. In this method, we obtain the junction tree with minimum state space and complete the inference. The experimental results show that the proposed algorithm help to improve convergence speed and obtain junction tree with lower state space values.