论文部分内容阅读
作者针对单个核函数构成的SVM并不能满足诸如数据异构或不规则、样本规模巨大、样本分布不平坦等实际应用的需求,而将多个核函数进行组合,以获得更好的效果,提出一种基于多核的模糊支持向量机算法。此算法决策树中的模糊核权重主要是借助于样本各自的模糊因子来确定。仿真实验数据表明:与传统单核函数支持向量机相比,多核模糊支持向量机具有很好的优越性。
The author poses that the SVM composed of a single kernel function can not satisfy the practical application such as data heterogeneity or irregularity, large sample size and uneven sample distribution, and the combination of multiple kernel functions can achieve better results. A Multi - Core Fuzzy Support Vector Machine Algorithm. The weight of the fuzzy kernel in the decision tree of this algorithm is mainly determined by means of the respective fuzzy factors of the samples. Simulation experimental data show that compared with the traditional single-kernel support vector machine, multi-core fuzzy support vector machine has good advantages.