Abstract: This paper proposes an Active Support Vector Machine(ASVM) algorithm based on similarity fusion. It uses unlabeled samples, labeled samples, combines with Support Vector Machine(SVM) method to achieve active learning. Experimental results show, compared with the general ASVM, it can reduce the number of labeled samples effectively and inhibit the isolated samples of impact on the premise of keeping correctness of the classifier and it also has a higher classification accuracy in the same number of labeled samples.

Key words: active learning, Support Vector Machine(SVM), similarity fusion, labeled sample

摘要： 提出一种基于相似度融合的主动支持向量机算法，利用未标记样本和标记样本，结合支持向量机的方法实现主动学习。实验结果表明，该算法与普通主动学习的支持向量机相比，在保证分类器性能的情况下，可以减少标记样本的数目，抑制孤立样本对分类器的影响；在相同标记样本数目的情况下，该算法具有较高的分类精度。

关键词: 主动学习, 支持向量机, 相似度融合, 标记样本

CLC Number: 

• TP181