Abstract: A new method of 2_a_2 SVM multi-class classification is put forward, the advantage of which is to make good use of the results of each sub-classifier so that the problem of classification of different classes can be resolved only by using the smallest numbers of sub-classifiers.In order to verify the effectiveness of this method，experiments have been made on CMU database，and the experimental results are satisfactory.If this method is applied to resolve the problem of classification of lots of classes，comparing with the current methods，it also has the same advantage.

Key words: Support Vector Machine（SVM）, multi-class classification, 2_a_2 method

摘要： 提出一种2_a_2支持向量机多类分类新方法，它的优点是充分利用了每个子分类器的识别结果，将最少数量的子分类器组合在一起，实现多类分类。通过对CMU表情库4种不同表情图像的分类识别实验表明，该算法能明显提高识别速率。将该方法应用于解决更多类的分类问题时，同样体现出优越性。

关键词: 支持向量机（SVM）, 多类分类, 2_a_2方法