论文部分内容阅读
为了应对BP神经网络在复杂样本分类过程中存在的因为网络结构复杂,网络跳转多样而导致的分类能力低,分类结果模糊等问题,提出了基于因子分析和遗传算法的BP神经网络。本算法首先通过因子分析的方法降低BP神经网输入样本的维度,然后使用遗传算法改进BP神经网络的分析过程,从而实现BP神经网络的强分类分析能力。
In order to cope with the problems of BP neural network in complicated sample classification, such as complex network structure, diversified network jumps, low classification ability and fuzzy classification results, a BP neural network based on factor analysis and genetic algorithm is proposed. Firstly, this method reduces the dimension of BP neural network input sample through factor analysis, and then uses genetic algorithm to improve the analysis process of BP neural network, so as to realize the strong classification analysis ability of BP neural network.