A diversity-guided heuristic-based genetic algorithm for triangulation of Bayesian networks
Xuchu Dong, Haihong Yu, Dantong Ouyang, Yuxin Ye, Yonggang Zhang · Networked Computing and Advanced Information Management · 2010
For the optimization problem about triangulation of Bayesian networks, a novel genetic algorithm, DHGA, is proposed in this paper. DHGA employs a heuristic-based mutation operation. Moreover, it uses population diversity to identify stagnation and convergence as well as to guide the search procedure. Experiments on representative benchmarks show that DHGA posses better performance and robustness than other swarm intelligence methods.