Abstract
In machine learning applications where multiple data sources present, it is desirable to effectively exploit the sources simultaneously to make better inferences. When each data source is presented as a graph, a common strategy is to combine the graphs, e.g. by taking the sum of their adjacency matrices, and then apply standard graph-based learning algorithms. In this paper, we take an alternative approach to this problem. Instead of performing the combination step, a graph-based learner is created on each graph and makes predictions independently. The method works in an iterative manner: labels predicted by some learners in each round are added to the labeled set and the models are retrained. By nature, the method is based on two popular semi-supervised learning approaches: bootstrapping and graph-based methods, to take their advantages. We evaluated the method on the gene function prediction problem with real biological datasets. Experiments show that our method significantly outperforms a standard graph-based algorithm and compares favorably with a state-of-the-art gene function prediction method.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Abney, S.: Bootstrapping. In: Proceedings of ACL 2002, pp. 360–367 (2002)
Argyriou, A., Herbster, M., Pontil, M.: Combining graph Laplacians for semi- supervised learning. In: Advances in Neural Information Processing Systems18, NIPS 18, MIT Press, Cambridge (2006)
Balcan, M.-F., Blum, A., Yang, K.: Co-training and expansion: Towards bridging theory and practice. In: NIPS 17, pp. 89–96 (2005)
Barutcuoglu, Z., Schapire, R., Troyanskaya, O.: Hierarchical multi-label prediction of gene function. Bioinformatics 22(7), 830–836 (2006)
Blum, A., Mitchel, T.: Combining labeled and unlabeled data with co-training. In: Proc. Workshop on on Computational Learning Theory, COLT 1998, pp. 92–100 (1998)
Chapelle, O., Scholkopf, B., Zien, A. (eds.): SemiSupervised Learning. MIT Press (2006)
Mostafavi, S., Ray, D., Warde-Farley, D., Grouios, C., Morris, Q.: GeneMANIA: a real-time multiple association network integration algorithm for predicting gene function. Genome Biology 9(suppl. 1), S4 (2008)
Mostafavi, S., Morris, Q.: Fast integration of heterogeneous data sources for predicting gene function with limited annotation. Bioinformatics 26(14), 1759–1765 (2010)
Muhlenbach, F., Lallich, S., Zighed, D.A.: Identifying and handling mislabeled instances. Journal of Intelligent Information Systems 22(1), 89–109 (2004)
Pavlidis, P., Weston, J., Cai, J., Grundy, W.N.: Gene functional classification from heterogeneous data. In: Proceedings of RECOMB 2001, pp. 249–255 (2001)
Tang, W., Lu, Z., Dhillon, I.S.: Clustering with multiple graphs. In: ICDM 2009, pp. 1016–1021 (2009)
Tsuda, K., Shin, H., SchÄolkopf, B.: Fast protein classification with multiple networks. Bioinformatics 21, ii59–ii65 (2005)
Wang, W., Zhou, Z.H.: A new analysis of co-training. In: Proceedings of International Conference on Machine Learning, ICML 2010, pp. 1135–1142 (2010)
Zhang, M.L., Zhou, Z.H.: CoTrade: Confident co-training with data editing. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics 99, 1–15 (2011)
Zhou, D., Bousquet, O., Lal, T., Weston, J., Scholkopf, B.: Learning with local and global consistency. In: Advances in Neural Information Processing Systems, vol. 16. MIT Press, Cambridge (2004)
Zhou, D., Burges, C.J.C.: Spectral clustering and transductive learning with multiple views. In: Proceedings of the 24th International Conference on Machine Learning, ICML 2007, pp. 1159–1166 (2007)
Zhou, Z.-H., Li, M.: Tri-training: Exploiting unlabeled data using three classifiers. IEEE Transactions on Knowledge and Data Engineering 17(11), 1529–1541 (2005)
Zhu, X.: Semi-supervised learning literature survey. Technical Report 1530, Department of Computer Science, University of Wisconsin at Madison (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nhung, N.P., Phuong, T.M. (2011). A Bootstrapping Method for Learning from Heterogeneous Data. In: Kim, Th., et al. Future Generation Information Technology. FGIT 2011. Lecture Notes in Computer Science, vol 7105. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-27142-7_49
Download citation
DOI: https://doi.org/10.1007/978-3-642-27142-7_49
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-27141-0
Online ISBN: 978-3-642-27142-7
eBook Packages: Computer ScienceComputer Science (R0)