|
|
Bayesian Network Structure Learning Based on Improved Hybrid Genetic Bacterial Foraging Algorithm |
LIU Hao-ran1,2,CHANG Jin-feng1,PANG Na-na1,LI Chen-ran1,LU Ze-dan1 |
1. Information Science and Engineering College of Yanshan University, Qinhuangdao, Hebei 066004, China
2. Hebei Province Key Laboratory of Special Optical Fiber & Optical Fiber Sensing, Qinhuangdao,Hebei 066004, China |
|
|
Abstract Aiming at the problem that learning Bayesian network by heuristic learning algorithm is easy to get into local optimization and low searching efficiency, a hybrid improved genetic bacteria foraging optimization algorithm for learning Bayesian network structure was proposed. Firstly, the optimal population was obtained by genetic algorithm and used as the initial population of bacteria foraging algorithm. Then, in order to increase population diversity and expand search space, crossover and mutation strategies were used to improve the replication behavior of bacteria foraging algorithm. Finally, the population was updated through the initialization operation of the migration behavior of the improved bacteria foraging algorithm to prevent the loss of elite individuals. Through the iterative search of the population, the optimal Bayesian network structure was finally obtained.The simulation results show that the convergence accuracy and efficiency of this algorithm are improved compared with other algorithms.
|
Received: 17 October 2019
Published: 28 August 2020
|
|
|
|
|
[1]Wang J, Liu S. Novel binary encoding water cycle algorithm for solving Bayesian network structures learning problem[J]. Knowledge-Based Systems, 2018, 150(3): 95-110.
[2]刘广怡, 李鸥, 宋涛, 等. 基于贝叶斯网络的无线传感网高效数据传输方法[J]. 电子与信息学报, 2016, 38(6): 1362-1367.
Liu G Y, Li O, Song T, et al. Energy-efficiency data transmission method in WSN based on Bayesian network[J]. Journal of Electronics & Information Technology, 2016, 38(6): 1362-1367.
[3]Xuan J, Lu J, Zhang G, et al. Bayesian nonparametric relational topic model through dependent gamma processes[J]. IEEE Transactions on Knowledge and Data Engineering, 2016, 29(7): 1357-1369.
[4]Villanueva E , Maciel C D . Novel binary encoding water cycle algorithm for solving Bayesian network structures learning problem[J]. Neurocomputing, 2014, 123(3): 3-12.
[5]黄解军, 万幼川, 潘和平. 贝叶斯网络结构学习及其应用研究[J]. 武汉大学学报: 信息科学版, 2004, 29(4): 315-318.
Huang J J, Wan Y C, Pan H P. Bayesian Network structure learning and its applications[J]. Geomatics and Information Science of Wuhan University, 2004, 29(4): 315-318.
[6]冀俊忠, 张鸿勋, 胡仁兵, 等. 基于蚁群算法的贝叶斯网结构学习[J]. 北京工业大学学报, 2011, 37(6): 933-939.
Ji J Z, Zhang H X, Hu R B, et al. Learning Bayesian network structure based on ant colony optimization algorithms[J]. Journal of Beijing University of Technology, 2011, 37(6): 933-939.
[7]汪春峰, 张永红. 基于无约束优化和遗传算法的贝叶斯网络结构学习方法[J]. 控制与决策, 2013, 28(4): 618-622.
Wang C F, Zhang Y H. Bayesian network structure learning based on unconstrained optimization and genetic algorithm[J]. Control and decision, 2013, 28(4): 618-622.
[8]张平, 刘三阳, 朱明敏.基于人工蜂群算法的贝叶斯网络结构学习[J]. 智能系统学报, 2014, 9(03): 325-329.
Zhang P, Liu S Y, Zhu M M. Structure learning of Bayesian networks by use of the artificial bee colony algorithm[J]. CAAI Transactions on Intelligent Systems, 2014, 9(03): 325-329.
[9]Gheisari S, Meybodi M R. BNC-PSO: structure learning of Bayesian networks by Particle Swarm Optimization[J]. Information Sciences, 2016, 348(1): 272-289.
[10]刘浩然, 孙美婷, 王海羽, 等. 基于分类优化贝叶斯结构算法的篦冷机参数状态分析及其算法收敛性分析[J]. 计量学报, 2019, 40(4): 662-669.
Liu H R, Sun M T, Wang H Y, et al. Parameter State Analysis of G rate Cooler Based on Bayesian Structure Algorithm Based on Classification Optimization and Convergence Analysis[J]. Acta Metrologica Sinica, 2019, 40(4): 662-669.
[11]Ji J, Yang C, Liu J, et al. A comparative study on swarm intelligence for structure learning of Bayesian networks[J]. Soft Computing, 2017, 21(22): 6713-6738.
[12]席裕庚, 柴天佑. 遗传算法综述[J]. 控制理论与应用, 1996, 13(6): 697-708.
Xi Y G, Chai T Y. Summary of genetic algorithms[J]. Control Theory and Applications, 1996, 13(6): 697-708.
[13]Chow C, Liu C. Approximating discrete probability distributions with dependence trees[J]. IEEE transactions on Information Theory, 1968, 14(3): 462-467.
[14]姜瑞, 陈晓怀, 王汉斌, 等.基于贝叶斯信息融合的测量不确定度评定与实时更新[J]. 计量学报, 2017, 38(1): 123-126.
Jiang R, Chen X H, Wang H B, et al. Evaluation and Real Time Updating of Measurement Uncertainty Based on Bayesian Information Fusion[J]. Acta Metrologica Sinica, 2017, 38(1): 123-126.
[15]杨文, 颜卫, 涂尚坦, 等. 基于贝叶斯信息准则的极化干涉 SAR 图像非监督分类[J]. 电子与信息学报, 2012, 34(11): 2628-2634.
Yang W, Yan W, Tu S T, et al. An Unsupervised Based Classification Method of POLINSAR Image Based on Bayesian Information Criterion[J]. Journal of Electronics and Information Technology, 2012, 34(11): 2628-2634. |
|
|
|