【摘要】 现有的贝叶斯网络结构学习方法需要大量可靠例子进行复杂的运算,具有低效率和可靠性,而在操作风险管理方面积累大量可靠的例子非常困难。针对问题和实际需求,基于变量之间基本依赖关系、结点之间基本结构、d-separation标准和依赖分析方法进行小样本贝叶斯网络结构学习,分别使用模拟和真实数据进行了实验和分析,结果显示,该方法能够有效地进行小样本数据的贝叶斯网络结构学习。更多还原

【Abstract】 At present,the methods of learning Bayesian networks structure need a large number of data with high quality.The algorithms have low efficiency and reliability.But it is very difficult to accumulate many reliable examples in operational risk management.In this paper,a new method of learning Bayesian networks structure from small data set is presented based on basic dependency relationship between variables,basic structure between nodes,d-separation criterion and dependency analysis method.It can effectively avoid the problems above. The experiments and analysis are done by using stimulant and real-world dataset.Experimental results show that this method can effectively learn Bayesian network structure from small data set.更多还原