一般模糊极小极大神经网络的改进研究

针对一般模糊极小极大神经网络在处理重叠超盒和包含超盒时,出现新的类而标识为未知类,进而无法达到聚类预期效果的问题,提出了通过超盒的收缩过程来加入新类或删除一个已存在类的一般模糊极小极大神经网络,它继承了一般模糊极小极大神经网络的优点,并且避免了一般模糊极小极大神经网络在分类时的随意性,弥补了一般模糊极小极大神经网络无法达到聚类预期效果的目的,以及提高了模式分类的准确性和高效性.最后,通过实例验证了方法实用有效.

Research on Improved General Fuzzy Min-max Neural Network

Through adding new class or deleting an existing class of general fuzzy min-max neural network(GFMN),a new GFMN is proposed.It inherits the merit of the GFMN.A new GFMN avoids randomness of the traditional GFMN in classification and makes up the purpose of the traditional GFMN can't reach clustering the desired effect.At the same time,it improves the accuracy and efficiency in the pattern classification.The experimental results demonstrate the effectiveness of this new approach.