模糊粗糙集方法在样本归一化中的应用

Application of Fuzzy Rough Set Theory to Sample Normalization

提出了基于模糊粗糙集理论的样本归一化方法,用于解决因神经网络分类器在不同类样本间距离较近时训练速度较慢的问题。将神经网络的输入作为粗糙集信息系统的条件属性,神经网络的输出作为决策属性,构建决策表。利用粗糙集理论对训练样本离散化,根据离散化样本与两类不同样本间的距离差和两类样本的能量差,利用模糊集理论对该原始样本进行伸缩处理,然后,对伸缩预处理后的样本进行归一化,最后,用归一化处理后的样本对神经网络进行训练。以配电网故障选线为例,对该方法进行了分析和验证。仿真实验结果表明,经模糊粗糙集理论样本归一化处理后的神经网络训练时间明显缩短。因此,该方法正确、有效。

A sample normalization algorithm based on fuzzy rough set theory is proposed to avoid the longtime training of neural network classifier caused by the smaller distances between samples of different classes.The inputs of neural network are taken as the condition attributes and the output as the decision attribute,and the information system is established.The samples are discretized by using rough set theory.According to the distance differences between the discretized samples and two class of samples and the energy differences between the two class of samples,the original samples are extended or contracted based on fuzzy set theory.The extended or contracted samples are normalized.The normalized samples are used to train the neural network.The method is analyzed and verified with an example of faulty line detection for distribution network.The simulation results show that the training time of neural network with preprocessed samples by using fuzzy rough set theory is shorter markedly.So the proposed algorithm is correct and effective.