利用模糊神经网络实现数值信息与语言信息的融合

提出一种数值信息与语言信息融合的实现方法，融合是通过一个模糊神经网络完成的。该方法可用于对既有语言型变量，又有数值型变量的系统建立模型。实现融合的关键是对语言变量定义隶属函数。给出一种考虑决策者的偏好来描述语言变量的隶属函数的方法，仿真结果验证了算法的有效性。

Realization of the Fusion of the Numerical and Linguistic Information Using a Fuzzy Neural Network

The problem on the fusion of the numerical information and the linguistic information is discussed. A method is presented to model the system, which has not only the numerical variable, but also the linguistic variable, such as the management systems, decision support systems etc. A four layer fuzzy neural network is used to realize the fusion. The key point of the fusion is to define the membership function of the linguistic variable. A method of determining the membership function of the linguistic variable concern with the preference of the decision person is proposed. The simulation result shows the validity of the presented method.