期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

基于异质信息融合的网络图像半监督学习方法 被引量：3

Web Image Semi-supervised Learning Method Based on Heterogeneous Information Fusion
下载PDF
导出
摘要 网络图像通常包含文本、颜色和纹理等异质信息.本文提出了一种基于多类异质信息融合的网络图像半监督学习方法—局部协同训练(Local co-training,LCT).该方法在每个视图(对应一类信息)上对每个样本点的邻域构建线性局部模型,利用一组局部模型来表示数据关系;基于信息传播和协同训练对模型进行增量式迭代更新.该算法在协同训练和基于图正则化的方法这两类半监督学习算法间建立了桥梁.局部协同训练算法能够准确地描述样本的复杂分布,并且可以进行高效的增量学习,有利于大规模网络图像的在线学习.在Corel,Pascal和ImageNet数据集上的实验结果表明该方法具有良好的性能. Web images generally consist of heterogeneous information including texts, colors and textures. This paper proposes a new method, called local co-training （LCT）, for semi-supervised classification of web images based on fusion of heterogeneous information. The proposed method employs a set of local linear models to represent data points of each view, and incrementally refines these models by exploiting unlabeled data with information propagation and co-training. The local co-training builds a bridge between graph-based methods and co-training. The local co-training can model the instance distribution accurately in the high-dimensional space, and learn local models incrementally, which benefits the online classification of large scale of web images. Experiments on Corel, Pascal and ImageNet datasets demonstrate that the local co-training can effectively improve the classification performance of learners by exploiting multiple attribute sets and unlabeled data.
作者 杜友田 李谦 周亚东 吴陈鹤
机构地区 西安交通大学智能网络与网络安全教育部重点实验室
出处 《自动化学报》 EI CSCD 北大核心 2012年第12期1923-1932,共10页 Acta Automatica Sinica
基金 国家自然科学基金(60905018 61202392 60921003 61175039) 十二五国家科技支撑重点项目(2011BAK08B02) 教育部博士点基金(20090201120032) 中央高校基本科研业务费(xjj2009041 2012jdhz08)资助~~
关键词 网络图像分类 异质信息 局部协同训练 机器学习 Web image classification, heterogeneous information, local co-training （LCT）, machine learning
分类号 TP391.41 [自动化与计算机技术—计算机应用技术] TP18 [自动化与计算机技术—控制理论与控制工程]
  • 相关文献

参考文献21

  • 1Wu Y, Chang E Y, Chang K C C, Smith J R. Optimal multimodal fusion for multimedia data analysis. In: Pro- ceedings of the 12th Annual ACM International Conference on Multimedia. New York, USA: ACM, 2004. 572-579.
  • 2Jiang T, Tan A H. Learning image-text associations. IEEE Transactions Knowledge and Data Engineering, 2009, 21 (2): 161-177.
  • 3Zhu X J. Semi-supervised learning literature survey. Com- puter Sciences TR 1530, University of Wisconsin - Madison, USA, 2008, 1-60.
  • 4Blum A, Mitchell T. Combining labeled and unlabeled data with co-training. In: Proceedings of the llth Annual Con- ference on Computational Learning Theory. Madison, Wis- consin, USA: ACM, 1998. 92-100.
  • 5Zhang M L, Zhou Z H. CoTrade: confident co-training with data editing. IEEE Transactions on Systems, Man, and Cy- bernetics - Part B: Cybernetics, 2011, 41(6): 1612-1626.
  • 6Yan R, Naphade M. Semi-supervised cross feature learning for semantic concept detection in videos. In: Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, USA: IEEE, 2005. 657-663.
  • 7Gupta S, Kim J, Grauman K, Mooney R. Watch, listen & learn: co-training on captioned images and videos. Lecture Notes in Computer Science, 2008, 5211:457-472.
  • 8Kumar A, Daum6 HIII. A co-training approach for multi- view spectral clustering. In: Proceedings of the 2011 Inter- national Conference on Machine Learning. Bellevue, Wash- ington, USA: ACM, 2011. 393-400.
  • 9Belkin M, Matveeva I, Niyogi P. Regularization and semi- supervised learning on large graphs. In: Proceedings of the 2004 International Conference on Computational Learning Theory. Banff, Canada: ACM, 2004. 624-638.
  • 10Goldman S A, Zhou Y. Enhancing supervised learning with unlabeled data. In: Proceedings of the 17th International Conference on Machine Learning. Stanford, USA: ACM, 2000. 327-334.

同被引文献33

  • 1LIU Xinhai, Glinzel W, de Moor B. Hybrid clustering of multi-view data via Tucker-2 model and its application[J]. Scientometrics, 2011, 88(3): 819-839.
  • 2WANG Xiang, QIAN Buyue, YE Jieping, et al. Multi-objective multi-view spectral clustering via pareto optimization[C]//Proc of the 13th SIAM International Conference on Data Mining. Philadelphia: SIAM, 2013: 234-242.
  • 3LONG Bo, YU P S, ZHANG Zhongfei. A general model for multiple view unsupervised learning[CJ//Proc of the 8th SIAM International Conference on Data Mining. Philadelphia: SIAM, 2008: 822-833.
  • 4Kumar A, Hal Daum6 IIL A co-training approach for multi-view spectral clustering[C]//Proc of the 28th IEEE International Conference on Machine Learning. NJ: IEEE, 201 l: 393-400.
  • 5Kumar A, Rai P, Hal Daum6 III. Co-regularized multi-view spectral clustering[C]//Proc of the 24th Annual Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2011: 1413-1421.
  • 6CHEN Xiaojun, XU Xiaofei, HUANG Joshua, et al. TW-k-means: Automated two-level variable weighting clustering algorithm for multiview data[J]. IEEE Transactions on Knowledge and Data Engineering, 2013, 25(4): 932-944.
  • 7Blaschko M B, Lampert C H. Correlational spectral clustering[C]//Proc of the 21st IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2008: 1-8.
  • 8Chaudhuri K, Kakade S, Livescu K, et al. Multiview clustering via canonical correlation analysis[C]//Proc of the 26th Annual International Conference on Machine Learning. New York: ACM, 2009: 129-136.
  • 9LIU Jialu, WANG Chi, GAO Jing, et al. Multi-view clustering via oint nonnegative matrix factorization[C]//Proc of the 13th SIAM International Conference on Data Mining. Piscataway: IEEE, 2013: 252-260.
  • 10Tzortzis G, Likas A. Kernel-based weighted multi-view clustering[C]//Proc of the 12th IEEE international conference on Data Mining. Piscataway: IEEE, 2012: 675-684.

引证文献3

二级引证文献45

自动化学报
2012年 第12期

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部