基于混合樽海鞘-差分进化算法的贝叶斯网络结构学习算法

doi:10.11959/j.issn.1000-436x.2019124

Abstract

Abstract:

Aiming at the disadvantages of Bayesian network structure learned by heuristic algorithms,which were trapping in local minimums and having low search efficiency,a method of learning Bayesian network structure based on hybrid binary slap swarm-differential evolution algorithm was proposed.An adaptive scale factor was used to balance local and global search in the swarm grouping stage.The improved mutation operator and crossover operator were taken into salp search strategy and differential search strategy respectively to renew different subswarms in the update stage.Two-point mutation operator was adopted to improve the swarm’s diversity in the stage of merging of subswarms.The convergence analysis of the proposed algorithm demonstrates that best structure can be found through the iterative search of population.Experimental results show that the convergence accuracy and efficiency of the proposed algorithm are improved compared with other algorithms.

Key words: Bayesian network structure learning, slap swarm algorithm, differential evolution algorithm, adaptive factor

CLC Number:

TP18

Bin LIU,Ruixing FAN,Haoran LIU,Liyue ZHANG,Haiyu WANG,Chunlan ZHANG. Bayesian network structure learning algorithm based on hybrid binary salp swarm-differential evolution algorithm[J]. Journal on Communications, 2019, 40(7): 151-161.

Figures/Tables 15

References 21

[1]	WANG J , LIU S . Novel binary encoding water cycle algorithm for solving bayesian network structures learning problem[J]. Knowledge-Based Systems, 2018,150: 95-110.
[2]	XUAN J , LU J , ZHANG G ,et al. Bayesian nonparametric relational topic model through dependent Gamma processes[J]. IEEE Transactions on Knowledge ＆ Data Engineering, 2017,29(7): 1357-1369.
[3]	耿杨, 邵苏杰, 郭少勇 ,等. 基于可见损伤持续时间贝叶斯网络的视频QoE评估方法[J]. 通信学报, 2017,38(6): 136-141.
	GENG Y , SHAO S J , GUO S Y ,et al. Bayesian network-based video QoE assessment method using image sustained damage analysis[J]. Journal on Communications, 2017,38(6): 136-141.
[4]	ALONSO J I , OSSA L D L , GáMEZ J A ,et al. On the use of local search heuristics to improve GES-based Bayesian network learning[J]. Applied Soft Computing, 2018,64: 366-376.
[5]	SCANAGATTA M , CORANI G , CAMPOS C P D ,et al. Approximate structure learning for large Bayesian networks[J]. Machine Learning, 2018,15: 1-19.
[6]	LIU H , ZHOU S , LAM W ,et al. A new hybrid method for learning bayesian networks:Separation and reunion[J]. Knowledge-Based Systems, 2017,121: 185-197.
[7]	YANG C , JI J , LIU J ,et al. Structural learning of Bayesian networks by bacterial foraging optimization[J]. International Journal of Approximate Reasoning, 2016,69(C): 147-167.
[8]	TSAMARDINOS I , BROWN L E , ALIFERIS C F . The max-min hill-climbing Bayesian network structure learning algorithm[J]. Machine Learning, 2006,65(1): 31-78.
[9]	刘浩然, 孙美婷, 李雷 ,等. 基于蚁群节点寻优的贝叶斯网络结构算法研究[J]. 仪器仪表学报, 2017,38(1): 143-150.
	LIU H R , SUN M T , LI L ,et al. Bayesian network structure algorithm based on ant colony node optimization[J]. Chinese Journal of Scientific Instrument, 2017,38(1): 143-150.
[10]	CONTALDI C , VAFAEE F , NELSON P C . Bayesian network hybrid learning using an elite-guided genetic algorithm[J]. Artificial Intelligence Review, 2018,293: 1-28.
[11]	SEYEDALL M , AMIR H G , SEYEDEH Z M ,et al. Salp swarm algorithm:a bio-inspired optimizer for engineering design problems[J]. Advances in Engineering Software, 2017,114(1): 163-191.
[12]	SAYED G I , KHORIBA G , HAGGAG M H . A novel chaotic salp swarm algorithm for global optimization and feature selection[J]. Applied Intelligence, 2018,48: 3462-3481
[13]	刘会宇, 韩继红, 袁霖 ,等. 基于双变异策略的自适应骨架差分进化算法[J]. 通信学报, 2017,38(8): 201-212.
	LIU H Y , HAN J H , YUAN L ,et al. Self-adaptive bare-bones differential evolution based on bi-mutation strategy[J]. Journal on Communications, 2017,38(8): 201-212.
[14]	WANG G G , GANDOMI A H , ALAVI A H ,et al. A hybrid method based on krill herd and quantum-behaved particle swarm optimization[J]. Neural Computing and Applications, 2016,27(4): 989-1006.
[15]	JIAO J , VENKAT K , WEISSMAN T . Mutual information,relative entropy and estimation error in semi-martingale channels[J]. IEEE Transactions on Information Theory, 2017,PP(99):1.
[16]	刘浩然, 吕晓贺, 李轩 ,等. 基于 Bayesian 改进算法的回转窑故障诊断模型研究[J]. 仪器仪表学报, 2015,36(7): 1554-1561.
	LIU H R , LV X H , LI X ,et al. A study of fault diagnosis model of rotary kiln based on improved algorithm of Bayesian[J]. Chinese Journal of Scientific Instrument, 2015,36(7): 1554-1561.
[17]	GOUDOS S . Antenna design using binary differential evolution:application to discrete-valued design problems[J]. IEEE Antennas ＆Propagation Magazine, 2017,59(1): 74-93.
[18]	SOLIS F , WETS R . Minimization by random search technologies[J]. Mathematics of Operations Research, 1981,6: 19-30.
[19]	ZAKHAROV V K , RODIONOV T V . Naturalness of the class of Lebesgue-Borel-Hausdorff measurable functions[J]. Mathematical Notes, 2014,95(3-4): 500-508.
[20]	陈志敏, 田梦楚, 吴盘龙 ,等. 基于蝙蝠算法的粒子滤波法研究[J]. 物理学报, 2017,66(5): 41-50.
	CHEN Z M , TIAN M C , WU P L ,et al. Research on particle filtering based on bat algorithm[J]. Journal of Physics, 2017,66(5): 41-50.
[21]	LIU F , ZHANG S W , GUO W F ,et al. Inference of gene regulatory network based on local bayesian networks[J]. Plos Computational Biology, 2016,12(8):e1005024.

Metrics

Recommended 0

No Suggested Reading articles found!

数据量/个		BIC评分
数据量/个	f_best	f_av	f_sn
500	-1 139.83	-1 143.31	-1 133.84
1 000	-2 243.91	-2 265.09	-2 243.91
3 000	-6 822.17	-6 825.32	-6 822.17
5 000	-11 228.53	-11 243.58	-11 228.53

数据量/个		BIC评分
数据量/个	f_best	f_av	f_sn
500	-3 387.12	-3 392.34	-3 357.18
1 000	-6 980.96	-7 083.95	-6 980.94
3 000	-19 610.25	-19 930.82	-19 610.25
5 000	-33 565.68	-33 965.16	-33 565.65

数据量/个	f_best	f_av	f_sn
500	-5 221.83	-5 310.02	-5 196.92
1 000	-10 148.52	-10 180.08	-10 017.97
3 000	-29 227.87	-29 780.76	-28 993.16
5 000	-48 228.58	-48 920.58	-47 753.58

数据量/个			BIC评分
数据量/个	HBSS-DE	MAK	MMHC	AESL-GA	BEWCA-BN
500	-1 143.31	-1 256.04	-1 285.34	-1 294.32	-1 143.32
1 000	-2 265.06	-2 653.03	-2 294.76	-2 310.48	-2 265.85
3 000	-6 825.32	-7 069.02	-6 898.79	-6 948.57	-6 894.61
5 000	-11 243.58	-12 109.55	-11 956.57	-11 789.65	-1 1678.23

数据量/个			BIC评分
数据量/个	HBSS-DE	MAK	MMHC	AESL-GA	BEWCA-BN
500	-3 392.34	-3 498.95	-3 405.89	-3 425.87	-3 398.72
1 000	-7 083.95	-7 393.02	-7 209.63	-7 298.65	-7 144.06
3 000	-19 930.82	-21 970.46	-20 446.78	-20 324.85	-20 881.64
5 000	-33 965.16	-35 461.82	-35 192.46	-34 780.25	-34 455.11

Bayesian network structure learning algorithm based on hybrid binary salp swarm-differential evolution algorithm

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 15

References 21

Related Articles 1

Metrics

Recommended 0

数据量/个			执行时间/s
数据量/个	HBSS-DE	MAK	MMHC	AESL-GA	BEWCA-BN
500	2.11	3.62	4.52	1.58	3.85
1 000	2.37	3.74	4.85	2.21	4.14
3 000	2.85	4.32	5.62	2.89	4.69
5 000	3.03	5.31	7.67	3.52	4.43

数据量/个			执行时间/s
数据量/个	HBSS-DE	MAK	MMHC	AESL-GA	BEWCA-BN
500	5.23	7.04	10.24	8.47	7.79
1 000	5.84	7.72	10.79	8.52	7.95
3 000	6.08	9.57	11.58	9.72	9.82
5 000	6.85	11.77	12.28	10.48	10.11

数据量/个			执行时间/s
数据量/个	HBSS-DE	MAK	MMHC	AESL-GA	BEWCA-BN
500	118.37	314.16	2 305.48	304.24	125.47
1 000	128.54	356.18	2 465.75	325.49	139.84
3 000	171.25	505.57	3 180.27	490.24	172.43
5 000	235.48	948.81	4 587.58	578.59	195.48