Abstract
We study classification when the majority of data is unlabeled, and only a small fraction is labeled: the so-called semi-supervised learning situation. Blum and Mitchell’s co-training is a popular semi-supervised algorithm [1] to use when we have multiple independent views of the entities to classify. An example of a multi-view situation is classifying web pages: one view may describe the pages by the words that occur on them, another view describes the pages by the words in the hyperlinks that point to them. In co-training two learners each form a model from the labeled data and then incrementally label small subsets of the unlabeled data for each other. The learners then re-estimate their model from the labeled data and the psuedo-labels provided by the learners. Though some analysis of the algorithm’s performance exists [1] the computation performed is still not well understood. We propose that each view in co-training is effectively performing incremental EM as postulated by Neal and Hinton [3], combined with a Bayesian classifier. This analysis suggests improvements over the core co-training algorithm. We introduce variations, which result in faster convergence to the maximum possible accuracy of classification than the core co-training algorithm, and therefore increase the learning efficiency. We empirically verify our claim for a number of data sets in the context of belief network learning.
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
Blum, A., Mitchell, T.: Combining Labeled and Unlabeled Data with Co- training. In: COLT (1998)
Davidson, I., Aminian, M.: Using the Central Limit Theorem for Belief Network Learning. In: Proceedings of the 8th Int. Symposium on A.I. and Math (2004)
Neal, M., Hinton, G.: A New View of the EM Algorithm that Justifies Incremental and Other Variants. In: Biometrika (1993)
Muslea, I., Minton, S., Knoblock, C.: Active + Semi-Supervised Learning = Robust Multi-View Learning. In: Proceedings of the 19th International Conference on Machine Learning, pp. 435–442 (2002)
Nigam, K., Ghani, R.: Understanding the Behavior of Co-training. In: Proceedings of the Workshop on Text Mining at the Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD (2000)
Seeger, M.: Learning with Labeled and Unlabeled Data. Technical Report (2000)
Bilmes, J.A.: A Gentle Tutorial of the EM Algorithm, TR-97-021 (April 1998)
Byrne, W., Gunawardana, A.: Comments on “Efficient Training Algorithm for HMM’s Using Incremental Estimation”. IEEE Transactions on Speech and Audio Processing 8(6) (November 2000)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Aminian, M. (2004). Co-training from an Incremental EM Perspective. In: Yang, Z.R., Yin, H., Everson, R.M. (eds) Intelligent Data Engineering and Automated Learning – IDEAL 2004. IDEAL 2004. Lecture Notes in Computer Science, vol 3177. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-28651-6_114
Download citation
DOI: https://doi.org/10.1007/978-3-540-28651-6_114
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-22881-3
Online ISBN: 978-3-540-28651-6
eBook Packages: Springer Book Archive