Abstract: This paper presents a novel direct constructing-based fuzzy multi-class support vector classifier，named as FCS-SVM，based on previous multi-class classification method by Crammer and Singer.This algorithm，the optimal problem and its constraints of multi-class classification are reconstructed，and its corresponding Lagrangian formula is re-deduced.Experimental comparison with the previous work indicates that the method can obtain better classification ratio.

摘要： 针对直接多类分类方法，提出了一种新的基于直接构造多类SVM分类器的模糊多类支持向量机算法FCS-SVM。在算法中，重构了优化问题及其约束条件，以及Lagrange公式，并进行了推导。通过在标准数据集上的几个实验，对这些算法进行了比较分析。实验结果表明提出的算法可以得到比较理想的分类精度。