自适应模糊辨识中关于模糊最近邻聚类学习算法的一种改进

An Improving Method on Fuzzy Nearest-neighbour \= Clustering Learning in Self-adaptive Fuzzy Identification

分析了一种基于模糊最近邻聚类的自适应模糊辨识方法，从模型简化的思想出发提出一种改进方法，并通过实例仿真证实了改进方法的优越性。最后引入Ｂｅｚｄｅｋ的聚类评价指标对两种聚类结果的有效性进行评判。

A kind of method on self-adaptive fuzzy identification based on fuzzy nearest-neighbour clustering is analyed, and an improving method from the view of model simplifying is presented. Simulation results shows its advantages. Finally, the clustering validity of them is judged according to Bezdeks index on clustering judgement, and the result shows the superiority and dependability of the improving method theoretically.