基于最小二乘支持向量机算法的南宋官窑出土瓷片分类

CLASSIFICATION OF ANCIENT CERAMIC PIECES FROM UNEARTHED OFFICIAL WARE AT HANGZHOU BASED ON LEAST SQUARE SUPPORT VECTOR MACHINE ALGORITHM

将最小二乘支持向量机(leastsquare support vector machine,LS-SVM)算法用于杭州南宋官窑2窑址出土瓷片的分类研究中,根据瓷片胎和釉的主要、次要和痕量元素组成对它们进行了分类,用留一法检验其分类效果,并与支持向量机(support vector machine,SVM)算法和自组织特征映射(self-organizing map,SOM)算法进行了比较。结果表明:SVM算法和LS-SVM算法比SOM算法更适合于处理"小样本"问题;一般情况下,SVM的分类效果比LS-SVM的分类效果好,但是LS-SVM具有更快的求解速度。

Ancient ceramic pieces for two types of official ware in the Southern Song Dynasty were classified by the least squares support vector machine （LS-SVM） algorithm according to the discrepancies in the major, minor and trace elements in the bodies and glazes. The classification effect was validated by the leave-one-out method and compared with the support vector machine （SVM） and self-organizing map （SOM） methods. The results show that the methods of SVM and LS-SVM are preferable to SOM for classifying small samples. Generally, SVM provides a more accurate classification than does LS-SVM; however, the calculation of LS-SVM is quicker than that of SVM when run in MATLAB.