摘要
提出了一种基于自组织映射(self-organizing map,SOM)解码的多类SVM算法。该方法首先按照纠错输出编码(error correcting output codes,ECOC)训练子SVM二分类器,然后根据训练样本的输出训练SOM网络,得到其最优权值,最后对未知数据进行分类,这样充分考虑到了二分类器的输出置信度,而且有效地克服了同时和多个类别的距离最小的情况。通过对实际的Iris数据和Yale人脸库的分类实验,结果表明,新算法对于解决多类SVM的分类问题是很有效的。
A multi-class SVM algorithm based on SOM decoding is presented. First, the binary SVM classifiers are trained according to the error correcting output codes （ECOC）. Then the SOM network is trained with the output of the training samples and the optimum weights are obtained. Finally the unknown data is classified. By this method, the confidence of the binary classifiers is completely considered with the case avoided that the same minimum distance to several classes is obtained. The experimental results on the Iris data set and Yale face database show that the new algorithm is feasible for the multi-class SVM.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2006年第9期1447-1450,共4页
Systems Engineering and Electronics
基金
国防预研基金资助课题(51407030103DZ0117)
关键词
多类支持向量机
解码算法
纠错输出编码
自组织映射
multi-class support vector machines
decoding method
error correcting output codes
self-organizing map
