迁移学习旨在利用大量已标签源域数据解决相关但不相同的目标域问题.当与某领域相关的新领域出现时,若重新标注新领域,则样本代价昂贵,丢弃所有旧领域数据又十分浪费.对此,基于SVM算法提出一种新颖的迁移学习算法—–TL-SVM,通过使用目标域少量已标签数据和大量相关领域的旧数据来为目标域构建一个高质量的分类模型,该方法既继承了基于经验风险最小化最大间隔SVM的优点,又弥补了传统SVM不能进行知识迁移的缺陷.实验结果验证了该算法的有效性. Migration learning aims to address a number of related but not identical target domain problems using a large number of labeled source domain data.When a new domain is associated with a domain, if the new domain is re-labeled, then the sample is expensive and discards all the old domain data Waste.Therefore, based on the SVM algorithm, a novel migration learning algorithm - TL-SVM is proposed, which builds a high-quality classification model for the target domain by using a small amount of tagged data in the target domain and a large amount of old data in related fields The method not only inherits the advantages of SVM based on minimization of empirical risk, but also makes up for the shortcomings that traditional SVM can not transfer knowledge.The experimental results verify the effectiveness of the proposed algorithm.