提出了一种基于集成学习的适于多类问题的神经网络集成模型,其基本模块由一个OAA方式的二类别分类器和一个互补多类分类器组成.测试表明,该模型在多类问题上比其他经典集成算法有着更高的精度,并且有较少存储空间和计算时间的优势. A neural network ensemble model based on integrated learning which is suitable for many kinds of problems is proposed.The basic module consists of a two-class OAA classifier and a complementary multi-class classifier.Experiments show that this model is applicable to many kinds of problems It has higher accuracy than other classical integration algorithms and has the advantage of less storage space and computation time.