Support-vector-machine-

  利用谱聚类方法在特 征向量空间中对原始样本数据进行重新表述使得在新表述中同一聚类中的样本能够更好地积聚在一起构建聚类核函数 并进而构造聚类核半监督支持向量机 使样本更好地满足半监督学习必须遵循的聚类假设 (Restated in the new formulation in the same cluster sample be better able to accumulate together to build the clustering of nuclear function and thus to construct the semi-supervised clustering of nuclear support vector method of spectral clustering in the feature vector space of the original sample data machine so that the sample to better meet the needs of semi-supervised learning clustering assumptions that must be followed)