作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (8): 153-154. doi: 10.3969/j.issn.1000-3428.2011.08.052

• 人工智能及识别技术 • 上一篇    下一篇

半监督局部判别分析

姜 伟 1,2,杨炳儒 1   

  1. (1. 北京科技大学信息工程学院,北京 100083;2. 辽宁师范大学数学学院,辽宁 大连 116029)
  • 出版日期:2011-04-20 发布日期:2012-10-31
  • 作者简介:姜 伟(1969-),男,副教授、博士研究生,主研方向:流形学习,模式识别;杨炳儒,教授、博士生导师
  • 基金资助:

    国家自然科学基金资助项目(60675030)

Semi-Supervised Local Discriminant Analysis

JIANG Wei 1,2, YANG Bing-ru 1   

  1. (1. School of Information Engineering, University of Science and Technology Beijing, Beijng 100083, China; 2. School of Mathematics, Liaoning Normal University, Dalian 116029, China)
  • Online:2011-04-20 Published:2012-10-31

摘要:

针对无监督学习及有监督学习算法的缺点,提出一种半监督局部判别分析的线性降维算法。数据在没有足够的训练样本时,局部结构比全局结构更重要。算法在每一个局部区域利用有标签数据推导出数据的局部判别结构,无标签数据和有标签数据推导出数据的内在几何结构。在ORL和Yale人脸数据库上的实验结果表明该算法是有效的。

关键词: 判别结构, 半监督, 局部保持投影, 局部判别分析

Abstract:

Aiming at the disadvantage of unsupervised method and supervised method, a linear dimensionality reduction method called Semi-supervised Local Discriminant Analysis(SLDA) is proposed. When there is no sufficient training sample, local structure is generally more important than global structure. SLDA utilizes the labeled data points to infer the local discriminant structure, as well as the intrinsic geometrical structure inferred from both labeled and unlabeled data points at each local area. Experimental results on ORL and Yale face recognition demonstrate the effectiveness of the algorithm.

Key words: discriminant structure, semi-supervised, Local Preserving Projection(LPP), local discriminant analysis

中图分类号: