论文部分内容阅读
提出了一种基于支持向量机学习的模糊分类器(FCBSVM)。介绍了FCBSVM的基本思想及其结构,分析了隶属函数参数和惩罚参数C对分类规则的产生以及分类性能的影响,并提出了参数确定方法。构建这种分类器时,先选用适当的隶属函数,构造核函数。然后,以训练模式作为中心,进行模糊划分,对每个模糊划分建立一条模糊IF-THEN分类规则。最后,利用支持向量机学习方法,求出支持向量和规则的参数。这种分类器将支持向量机和模糊集合理论的优点结合起来,实现了模糊划分和模糊分类规则的自动产生。用双螺旋线数据和典型的数据集对分类器的性能进行了实验评测,验证了分类器的有效性。
A fuzzy classifier based on support vector machine learning (FCBSVM) is proposed. The basic idea of FCBSVM and its structure are introduced. The influence of membership function parameter and penalty parameter C on the generation and classification performance of classification rules is analyzed. A method of parameter determination is proposed. When constructing this classifier, we choose the appropriate membership function to construct the kernel function. Then, taking the training pattern as the center, we make a fuzzy division and establish a fuzzy IF-THEN classification rule for each fuzzy division. Finally, using support vector machine learning method, find the parameters of support vector and rules. This classifier combines the advantages of SVM and fuzzy set theory to realize the automatic generation of fuzzy classification and fuzzy classification rules. The performance of the classifier was experimentally evaluated with double helix data and a typical data set to verify the validity of the classifier.