CAJ | 학술논문

从数据集中学习贝叶斯网络结构是一个NP难问题。针对此问题提出基于遗传算子的人工蜂群算法。首先，将贝叶斯网络结构映射为一种二进制编码；其次，根据贝叶斯网络的结构特点，设计了蜜源的更新策略，从而将学习贝叶斯网络结构的过程转化为蜂群寻找最优蜜源的过程。实验结果表明，该算法应用于贝叶斯网络结构学习中的有效性。
종수거집중학습패협사망락결구시일개NP난문제。침대차문제제출기우유전산자적인공봉군산법。수선，장패협사망락결구영사위일충이진제편마；기차，근거패협사망락적결구특점，설계료밀원적경신책략，종이장학습패협사망락결구적과정전화위봉군심조최우밀원적과정。실험결과표명，해산법응용우패협사망락결구학습중적유효성。
The learning structure of Bayesian networks from a data set is an NP-hard problem .To deal with this problem , an artificial bee colony algorithm based on genetic operators is proposed in this paper .The structure of the Bayesian network is mapped to binary encoding , and the updated strategy of nectar is designed according to the characteristics of the Bayesian network structure .Thus the process of structure learning of the Bayesian network is transformed into the process of the bee colony finding the optimal nectar .The experimental results show that the al-gorithm is valid in the structure learning of Bayesian networks .