Xiukuan Zhao
Key Laboratory of Ionospheric Environment, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, 100029, China
Baiqi Ning
Key Laboratory of Ionospheric Environment, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, 100029, China
Gangbing Song
Department of Mechanical Engineering, University of Houston, Houston, TX 77204, United State of America
ABSTRACT
Conventional artificial intelligent methods such as neural network and SVM use only labeled data (feature/label pairs) for training. Labeled instances are often difficult, expensive, or time consuming to obtain. To use a large amount of unlabeled data together with labeled data to build better models, we proposed an active semi-supervised learning method based on the pool query active learning and manifold regularization semi-supervised method. In this paper, the effectiveness of the method was verified by its application to a synthetic data set and three real world classification problems. The experimental results showed that employing our active semi-supervised learning method can significantly reduce the need for labeled training instances.
PDF References Citation
How to cite this article
Xiukuan Zhao, Baiqi Ning and Gangbing Song, 2013. An Effective Active Semi-supervised Learning Method Based on Manifold Regularization. Journal of Applied Sciences, 13: 1728-1733.
DOI: 10.3923/jas.2013.1728.1733
URL: https://scialert.net/abstract/?doi=jas.2013.1728.1733
DOI: 10.3923/jas.2013.1728.1733
URL: https://scialert.net/abstract/?doi=jas.2013.1728.1733
REFERENCES
- Belkin, M., P. Niyogi and V. Sindhwani, 2006. Manifold regularization: A geometric framework for learning from labeled and unlabeled examples. J. Mach. Learn. Res., 12: 2399-2434.
Direct Link - Chapelle, O., V. Sindhwani and S.S. Keerthi, 2008. Optimization techniques for semi-supervised support vector machines. J. Mach. Learn. Res., 9: 203-233.
Direct Link - Liu, Y., 2004. Active learning with support vector machine applied to gene expression data for cancer classification. J. Chem. Inform. Comput. Sci., 44: 1936-1941.
CrossRefDirect Link - Shen, J., B. Ju, T. Jiang, J. Ren, M. Zheng, C. Yao and L. Li, 2011. Column subset selection for active learning in image classification. Neurocomputing, 74: 3785-3792.
CrossRefDirect Link - Tong, S. and D. Koller, 2001. Support vector machine active learning with applications to text classification. J. Mach. Learn. Res., 2: 45-66.
Direct Link - Wu, J., Y.B. Diao, M.L. Li, Y.P. Fang and D.C. Ma, 2009. A semi-supervised learning based method: Laplacian support vector machine used in diabetes disease diagnosis. Interdiscip. Sci. Comput. Life Sci., 1: 151-155.
CrossRefPubMedDirect Link - Yu, D., B. Varadarajan, L. Deng and A. Acero, 2010. Active learning and semi-supervised learning for speech recognition: A unified framework using the global entropy reduction maximization criterion. Comput. Speech Language, 24: 433-444.
CrossRef - Zhang, C. and T. Chen, 2002. An active learning framework for content-based information retrieval. IEEE Trans. Multimedia, 4: 260-268.
CrossRef - Zhao, X., M. Li, J. Xu and G. Song, 2011. An effective procedure exploiting unlabeled data to build monitoring system. Expert Syst. Appl., 38: 10199-10204.
CrossRef - Zhao, X., X. Li, C. Pang and S. Wang, 2013. Human action recognition based on semi-supervised discriminant analysis with global constraint. Neurocomputing, 105: 45-50.
CrossRef - Zha, Z.J., M. Wang, Y.T. Zheng, Y. Yang, R. Hong and T.S. Chua, 2012. Interactive video indexing with statistical active learning. IEEE Trans. Multimedia, 14: 17-27.
CrossRefDirect Link - Zhu, X., J. Lafferty and Z. Ghahramani, 2003. Combining active learning and semi-supervised learning using gaussian fields and harmonic functions. Proceedings of the ICML 2003 Workshop on The Continuum from Labeled to Unlabeled Data in Machine Learning and Data Mining, August 2003, Washington, DC., pp: 58-65.