提出一种概率神经网络样例选择算法,它包括两个阶段,第一个阶段利用概率神经网络计算样例的不确定性,第二个阶段利用计算出的不确定性选择样例.与压缩近邻规则、编辑近邻规则、约简近邻规则和迭代过滤算法四种代表性的样例选择算法进行了实验比较,实验结果显示在能力保持框架下,该算法的性能优于这四种方法.本文提出的算法具有下列特点:(1)学习速度快;(2)没有分类器的限制;(3)具有好的泛化能力. This paper presents a sample selection algorithm of probabilistic neural network, which includes two phases, the first one uses the probabilistic neural network to calculate the uncertainty of the sample, the second one uses the calculated uncertainty to select a sample, Neighbor rule, edit neighbor rule, reduction neighborhood rule and iterative filtering algorithm. The experimental results show that the performance of this algorithm is better than those of the four methods in the context of ability preservation. The proposed algorithm has the following features: (1) fast learning speed; (2) no classifier restrictions; (3) good generalization ability.