摘要
投影双子支持向量机(PTSVM)是一种有监督学习方法,其性能极大依赖于有监督信息量的大小.受流形正则化框架启发,文中提出半监督投影双子支持向量机(SPTSVM).该方法可同时利用有监督(有标签样本)信息和无监督(无标签样本)信息构造一个更合理的半监督学习器.SPTSVM不仅继承PTSVM有监督分类性能,而且使用流形正则项捕获蕴含在无标签数据中的潜在几何信息.通过选择合理的参数,SPTSVM退化为有监督PTSVM或正则化PTSVM.在人工数据集和实际数据集上的对比实验验证文中方法的有效性.
Projection twin support vector machine (PTSVM) is a supervised learning method and its performance deteriorates when supervised information is insufficient. To resolve this issue, a semi-supervised projection twin support vector machine (SPTSVM) is proposed inspired by the manifold regularization. Both supervised (labeled) and unsupervised (unlabeled) information are utilized to build a more reasonable semi-supervised classifier. Compared with PTSVM, SPTSVM takes the intrinsic geometric information into full consideration via manifold regularization. Furthermore, by selecting appropriate parameters, SPTSVM degenerates into either supervised PTSVM or projection twin support vector machine with regularization term. The effectiveness of the proposed approach is demonstrated by comparison on both artificial and real-world datasets.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2016年第2期97-107,共11页
Pattern Recognition and Artificial Intelligence
基金
国家自然科学基金项目(No.11426202
11426200
61304125
11201426)
浙江省自然科学基金项目(No.LY15F030013
LQ13F030010)
浙江省教育厅科研基金项目(No.Y201225179)资助~~
关键词
半监督学习
支持向量机
投影双子支持向量机(PTSVM)
流形正则化
非平行投影
Semi-supervised Learning, Support Vector Machine, Projection Twin Support Vector Machine (PTSVM) , Manifold Regularization, Nonparallel Projection