Abstract
By only selecting the most informative instances for labeling, active learning could reduce the labeling cost when labeled instances are hard to obtain. Facing the same situation, semi-supervised learning utilize unlabeled instances to strengthen classifiers trained on labeled instances under suitable assumptions. However, the current active learning methods often ignore such effect. Combining semi-supervised learning, we propose a graph-based active learning method, which can also handle multi-class problems, in the entropy reduction framework. The proposed method trains the base classifier using a popular graph-based semi-supervised label propagation method and samples the instance with the largest expected entropy reduction for labeling. The experiments show that the proposed method outperforms the traditional sampling methods on selected datasets.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Hieu, T.N., Arnold, S.: Active learning using pre-clustering. In: Proc. 21th International Conf on Machine Learning, Banff, CA. Morgan Kaufmann, San Francisco (2004)
Muslea, I., Minton, S., Knoblock, C.A.: Active learning with multiple views. Journal of Artificial Intelligence Research 27, 203–233 (2006)
Lewis, D.D., Gale, W.A.: A sequential algorithm for training text classifiers. In: 17th ACM International Conference on Research and Development in Information Retrieval, pp. 3–12. Springer, Heidelberg (1994)
Tong, S., Koller, D.: Support vector machine active learning with applications to text classification. Journal of Machine Learning Research 2, 45–66 (2001)
Schohn, G., Cohn, D.: Less is more: Active learning with support vector machines. In: Proc. 17th International Conf. on Machine Learning, CA, pp. 839–846. Morgan Kaufmann, San Francisco (2000)
Cohn, D.A., Ghahramani, Z., Jordan, M.I.: Active learning with statistical models. Journal of Artificial Intelligence research 4, 129–145 (1996)
Roy, N., McCallum, A.: Toward optimal active learning through sampling estimation of error reduction. In: Proc. 18th International Conf. on Machine Learning, pp. 441–448. Morgan Kaufmann, San Francisco (2001)
Zhu, X., Lafferty, J.D., Ghahramani, Z.: Combining active learning and semi-supervised learning using gaussian fields and harmonic functions. In: Proceedings of the ICML 2003 Workshop on The Continuum from Labeled to Unlabeled Data (2003)
Seung, H.S., Opper, M., Sompolinsky, H.: Query by committee. In: Proceedings of the Fifth Workshop on Computational Learning Theory, San Mateo, CA, pp. 287–294. Morgan Kaufmann, San Francisco (1992)
Freund, Y., Seung, H.S., Shamir, E., Tishby, N.: Selective sampling using the query by committee algorithm. Machine Learning 28, 133–168 (1997)
Abe, N., Mamitsuka, H.: Query learning using boosting and bagging. In: Proc. 15th International Conf. on Machine Learning, Madison, CA, pp. 1–10. Morgan Kaufmann, San Francisco (1998)
Melville, P., Mooney, R.J.: Diverse ensembles for active learning. In: Proc. 21th International Conf. on Machine Learning, Banff, CA, pp. 584–591. Morgan Kaufmann, San Francisco (2004)
Long, J., Yin, J., Zhu, E.: An active learning method based on most possible misclassification sampling using committee. In: Torra, V., Narukawa, Y., Yoshida, Y. (eds.) MDAI 2007. LNCS (LNAI), vol. 4617, pp. 104–113. Springer, Heidelberg (2007)
McCallum, A., Nigam, K.: Employing em and pool-based active learning for text classification. In: ICML, pp. 350–358 (1998)
Zhu, X., Ghahramani, Z., Lafferty, J.D.: Semi-supervised learning using gaussian fields and harmonic functions. In: ICML, pp. 912–919 (2003)
Zhou, D., Bousquet, O., Lal, T.N., Weston, J., Schölkopf, B.: Learning with local and global consistency. In: NIPS (2003)
Zhu, X.: Semi-supervised learning literature survey (2006)
Blum, A., Chawla, S.: Learning from labeled and unlabeled data using graph mincuts. In: ICML, pp. 19–26 (2001)
Zhu, X., Ghahramani, Z.: Learning from labeled and unlabeled data with label propagation (2002)
Belkin, M., Niyogi, P.: Using manifold stucture for partially labeled classification. In: NIPS, pp. 929–936 (2002)
Chapelle, O., Schölkopf, B., Zien, A. (eds.): Semi-Supervised Learning. MIT Press, Cambridge (2006)
Baram, Y., El-Yaniv, R., Luz, K.: Online choice of active learning algorithms. In: ICML, pp. 19–26 (2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Long, J., Yin, J., Zhao, W., Zhu, E. (2008). Graph-Based Active Learning Based on Label Propagation. In: Torra, V., Narukawa, Y. (eds) Modeling Decisions for Artificial Intelligence. MDAI 2008. Lecture Notes in Computer Science(), vol 5285. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-88269-5_17
Download citation
DOI: https://doi.org/10.1007/978-3-540-88269-5_17
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-88268-8
Online ISBN: 978-3-540-88269-5
eBook Packages: Computer ScienceComputer Science (R0)