基于数值型数据的模糊规则快速挖掘方法  

Fast Generation Method of Fuzzy Rules from Numerical Data

作　　者：谢永芳[1] 胡志坤[2] 桂卫华[1] 

机构地区：[1]中南大学信息科学与工程学院,湖南长沙410083 [2]中南大学物理科学与技术学院,湖南长沙410083

出　　处：《控制工程》2006年第5期442-444,448,共4页Control Engineering of China

基　　金：国家自然科学基金资助项目(50374079);国家973资助项目(2002cb31200)

摘　　要：针对数值型数据能准确反应现实世界,但难以理解的问题,为了从数值型数据中挖掘出易于理解的知识,提出了基于数值型数据的模糊规则快速挖掘方法。该方法能从数值型数据中挖掘出一个零阶的Sugeno模糊规则,并采用一种启发式方法将这个零阶的Sugeno模糊规则的数值结论转变为两个带置信度的语言变量,并给出了规则库的存储结构。最后通过实例证明了这种快速模糊规则挖掘方法能避免复杂的数值型计算和能有效逼近非线性函数的优点。To the problem that the numerical data reflecting the real world is difficult to be understood, a fast generation method of fuzzy rules is pn^sented in order to extract the linguistic knowledge from numerical data. The fuzzy if-then rules with consequent real number are extracted from numerical data. A linguistic representation method for deriving linguistic rules from fuzzy if-then rules with consequent real numbers is suggested. The linguistic representation consists of two linguistic variables with the degree of certainty. The storage structure of rule base is described. Simulations show that the method involves neither time-consuming iterative learning procedures nor complicated rule generation mechanisms, and can approximate complex systems.

关 键 词：数值型数据 模糊规则 数据挖掘 Sugeno模型 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]