中文说明：
非平衡数据集的分类问题经常出现在许多实际应用中.支持向量机在处理这一类问题时,整体分类性能比较低.为此,Veropoulos提出的采用不同惩罚系数的改进算法可以较好的解决此类问题.此外,可以利用序列最小优化算法简单快速的解决上述优化问题.
English Description：
The classification of unbalanced data sets often appears in many practical applications When dealing with this kind of problems, the overall classification performance of support vector machine is relatively low Therefore, the improved algorithm with different penalty coefficients proposed by veropoulos can better solve this kind of problem In addition, the sequence minimum optimization algorithm can be used to solve the above optimization problems simply and quickly