可变模糊聚类及模式识别统一理论与模型

Unified theory and model of variable fuzzy clustering and pattern recognition

提出了可变模糊聚类及模式识别的统一理论与模型.统一模型的特点是可变模糊模式识别模型为可变模糊聚类模型的特例.给出了可变模糊模式识别模型的4种参数组合:(1)α=1,p=1;(2)α=1,p=2;(3)α=2,p=1;(4)α=2,p=2.给出了c=2时模糊模式识别4种优选决策模型之间的关系.统一理论与模型可以提高模糊聚类、模式识别、优选决策的可靠性.

The unified theory and model of variable fuzzy clustering and pattern recognition are proposed, and it is pointed out that variable fuzzy pattern recognition is a special case of variable fuzzy clustering. Then, the four combination forms of the parameters of variable fuzzy pattern recognition model are given. They are （1）α= 1,p= 1; （2）α= 1,p=2; （3）α=2,p= 1 and （4）α=2,p=2. At the same time, the relationship among the four optimization models of fuzzy pattern recognition is presented. The unified model proposed can improve the reliability degree of fuzzy clustering, pattern recognition and optimization decision.