中文摘要：

半监督学习是人工智能研究领域中的重要课题，结合有监督学习和无监督学习的优点来提高学习器的性能。针对有监督分类和无监督分类不能充分利用已标记样本和未标记样本的问题，介绍了半监督分类方法及其基本思想、研究现状、应用领域与常用算法，分析了当前半监督分类算法研究中的主要困难，同时提出了需要进一步研究的若干问题。

英文摘要：

Semi-supervised learning is an important topic in research field of artificial intelligence, it combines the advantages of supervised learning and unsupervised learning to improve the performance of learning machine. In view of the problem that supervised classification and unsupervised classification cannot make full use of labelled samples and unlabeled samples, we introduced the semi- supervised classification methods including the basic idea, status of research, application fields and commonly used algorithms and analysed the main difficulties in current study of semi-supervised classification algorithm, at the same time we put forward some problems that need to be studied further.