基于模糊聚类的学习样本选取方法

Method on pattern selection based on fuzzy clustering

针对复杂环境下神经网络学习效率较低的问题 ,提出了一种按模糊聚类筛选样本模式类别的矫正方法 .先将模式大类细分为具有相近指标值的子类 ,以子类中心为典型学习样本 ,构成新的学习样本集 ,并以此样本集训练神经网络 .实际应用证明 ,该方法在样本分布存在多峰性和交遇性的情况下 。

Considering the low\|efficiency of learning of neural network raised in complex conditions, we bring forward a correction strategy which selects the categories of sample by course of fuzzy clustering. Firstly, divide the main class into subclasses which assume the similar index value, take the subclass center as the typical learning sample then make a new set of learning sample, and train the neural network using this sample set. Experiments prove that this method is of service in upraising the learning efficiency of network when multimodality and admix exist in sample distribution.