Structure Learning Algorithm of DBN Based on Particle Swarm Optimization

According to the characteristics dynamic Bayesian network structure, in which the build process, due to the transfer of network nodes is twice the number of variables to be solved, using the traditional method of Bayesian network construction slow efficiency, this article will apply PSO dynamic Bayesian network structure learning, and According to the characteristics Transferred network node in the network is divided into two parts, using stepwise build network has to choose sides proposed DBN structure learning algorithm based on particle swarm optimization. The last benchmark datasets large number of experiments show that the algorithm can improve the efficiency of dynamic Bayesian network structure learning.