摘要
Bayesian网络学习的一种方法是根据输入数据集使用某种打分机制找到与数据集相拟合的候选网络.ACOB算法(蚁群优化B算法)是其中一种基于元启发引入蚂蚁机制来进行Bayesian网络学习的方法.本文在该算法基础之上提出一种改进算法--PACOB,并行进行Bayesian网络学习.实验结果表明,该并行算法相对于其串行算法具有一定的优势,提供了一种Bayesian网络学习问题的有效手段.
One of important approaches to learn Bayesian networks uses a scoring metric to find the most appropriate candidate network for the data base. ACOB（ant colony optimization B algorithm） is an algorithm of the metaheuristic to solve the prob- lem. An improved algorithm-PACOB is proposed which is based on ACOB. It shows a good performance compared with ACOB based on the experiments ,and it is one of good and forceful methods to learn Bayesian networks.
出处
《小型微型计算机系统》
CSCD
北大核心
2007年第4期651-655,共5页
Journal of Chinese Computer Systems
