An improved structure learning algorithm of Bayesian Network based on the hesitant fuzzy information flow

摘要: Abstract The Bayesian Network (BN) is one of the most effective theoretical models in fields uncertain reasoning. With nonlinear evolution events and complexity practical problems, there will be massive data with uncertainty, bringing more challenges to application BN. In this paper, by combining advantages hesitant fuzzy set (HFS) depicting information flow (IF) causal analysis systems, an improved Particle Swarm Optimization (PSO) algorithm for structure learning BN based on (HFIF) proposed. First, a new physical notion called HFIF defined depict relationship between two intensive variable sequences. Then global conducted. By constructing unconstrained optimization model, initial optimized search space significant causality are obtained, which, approximate optimal PSO directions arcs determined at same time. A specific implementation process under environment also presented. Moreover, proposed applied ASIA network BOBLO network. Comparisons traditional algorithms conducted demonstrate effectiveness environment.