| 支持向量机与模糊规则模型的等价性 |
| |
| 引用本文： | 彭新俊,彭中梅,胡光华.支持向量机与模糊规则模型的等价性[J].云南民族大学学报(自然科学版),2005,14(2):153-158. |
| |
| 作者姓名： | 彭新俊 彭中梅 胡光华 |
| |
| 作者单位： | 1. 云南大学,数学系,云南,昆明,650091
2. 嘉兴市大桥中学,浙江,嘉兴,314006 |
| |
| 摘 要： | 在模糊基函数为高斯型隶属函数或更一般地其满足Mercer条件和核函数为有界函数的情况下,证明了支持向量机器问题与一般的模糊规则模型的等价性.这一结论在许多实际复杂的无法事先确定其模糊规则的数量的情况下十分重要.并且给出了当在知道模糊模型时分别确定C和ε值的算法.最后用两个例子说明二者的等价性.
|
| 关 键 词： | 支持向量机 模糊模型 模糊基函数 核函数 |
The Equivalence of Support Vector Machine and Fuzzy Rule- Based Modeling |
| |
| PENG Xin-jun,PENG Zhong-mei,HU Guang-hua.The Equivalence of Support Vector Machine and Fuzzy Rule- Based Modeling[J].Journal of Yunnan Nationalities University:Natural Sciences Edition,2005,14(2):153-158. |
| |
| Authors: | PENG Xin-jun PENG Zhong-mei HU Guang-hua |
| |
| Institution: | PENG Xin-jun~1,PENG Zhong-mei~2,HU Guang-hua~1 |
| |
| Abstract: | The equivalence of the support vector machine(SVM) and the Fuzzy rule- based modeling (FRM) were proved under the condition that the fuzzy basis function(FBF) is the Gaussian membership function, or more generally, the fuzzy basis function satisfies the Mercer's condition and the maximum value of the kernel function is finite. This conclusion is important to many practical and complicated situations where one unable to determine the number of rules in advance. Moreover, the algorithms to determine C and ε are considered when the fuzzy modeling is known. Two examples are given in order to explain this equivalence. |
| |
| Keywords: | support vector machine fuzzy modeling fuzzy basis function kernel function |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
