基于自组织模糊神经网络的市场占有率模型

针对现有市场占有率模型(如传统的统计方法和近年来出现的模糊模型与神经网络模型)实际预测效果并不理想的问题，结合模糊神经网络和自组织理论建立了基于自组织模糊神经网络的市场占有率预测模型．该模型无需事先确定模糊控制规则，并能通过神经网络的结构及参数学习调整模糊神经网络的结构．最后结合实例进行了仿真．结果显示该模型优于其它模型，能为企业提供有力的市场占有率定量分析工具．

A Market Share Model Based on the Self-organized Fuzzy Neural Networks

Aiming at the unperfect forecast results obtained by the existing market share models, such as the traditional statistical methods and the fuzzy models or neural network models appeared in recent years, a market share predicting model based on the self-organized fuzzy neural network was presented according to the fuzzy neural networks and the theory of self-organization. By using this model, people need not select any fuzzy logic in advance, and can adjust the network structure by the structure and parameter learning of the neural network. A simulation for the proposed model was finally carried out with an example. The results show that the proposed model is better than other existing models, and is a powerful quantitative tool to analyze the market share for enterprises.