中文摘要：

同时利用半监督学习中的标记数据与非标记数据，可以提高机器学习的能力．为了利用非标记数据，我们将非标记数据所在的样本空间映射到标记样本空间．讨论如何利用非标记数据构造学习算法，并证明所构造算法的收敛性．

英文摘要：

The machine learning accuracy can be improved by making use of both labeled and unlabeled data in semi-supervised learning. In order to use unlabeled data, assumed both labeled and unlabeled data lie on a same manifold, and proposed an algorithm which mapped unlabeled data space into labeled data space, and proved this algorithm was corvengence.