Abstract
This paper proposes instance decomposition schemes (IDSs) for mapping multi-class classification tasks into a series of binary classification tasks. It demonstrates theoretically that IDSs can handle two main problems of the class decomposition schemes: the problem of difficult binary classification tasks and the problem of positive error correlation of the binary classifiers. The experiments show that IDSs can outperform standard ECOC class decompositions.
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
Notes
- 1.
Ripper was originally proposed as a binary classifier in [3].
- 2.
minNumObj is the minimal number of training instances to create a Ripper rule. Increasing the value of minNumObj decreases the Ripper complexity.
References
Alpaydin, E., Mayoraz, E.: Learning error-correcting output codes from data. In: Proceedings of the Ninth International Conference on Artificial Neural Networks (ICANN-99), pp. 743–748. MIT Press (1999)
Bache, K., Lichman, M.: UCI machine learning repository (2013)
Cohen, W.W.: Fast effective rule induction. In: Proceeding of the Twelfth International Conference on Machine Learning, pp. 115–123. Morgan Kaufmann (1995)
Dekel, O., Singer, Y.: Multiclass learning by probabilistic embeddings. Adv. Neural Inf. Process. Syst. 15, 945–952 (2002)
Dietterich, T.G., Bakiri, G.: Solving multiclass learning problems via error-correcting output codes. J. Artif. Intell. Res. 2, 263–286 (1995)
Escalera, S., Pujol, O., Radeva, P.: Error-correcting ouput codes library. J. Mach. Learn. Res. 11, 661–664 (2010)
Guyon, I., Elisseeff, A.: An introduction to variable and feature selection. J. Mach. Learn. Res. 3, 1157–1182 (2003)
Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The WEKA data mining software: an update. SIGKDD Explor. 11, 10–18 (2009)
le Cessie, S., van Houwelingen, J.C.: Ridge estimators in logistic regression. Appl. Stat. 41(1), 191–201 (1992)
Lorena, A.C., de Carvalho, A., Gama, J.M.P.: A review on the combination of binary classifiers in multiclass problems. Artif. Intell. Rev. 30, 19–37 (2008)
Marchiori, E.: Hit miss networks with applications to instance selection. J. Mach. Learn. Res. 9, 997–1017 (2008)
Nadeau, C., Bengio, Y.: Inference for the generalization error. In: Solla, S.S., Leen, T.K., Müller, K.-R. (eds.) Advances in Neural Information Processing Systems 12, pp. 307–313. The MIT Press, Cambridge (1999)
Pujol, O., Radeva, P., Vitrià , J.: Discriminant ECOC: a heuristic method for application dependent design of error correcting output codes. IEEE Trans. Pattern Anal. Mach. Intell. 28(6), 1007–1012 (2006)
Rätsch, G., Smola, A.J., Mika, S.: Adapting codes and embeddings for polychotomies. Adv. Neural Inf. Process. Syst. 15, 513–520 (2002)
Zhou, J., Peng, H., Suen, C.Y.: Data-driven decomposition for multi-class classification. Pattern Recogn. 41(1), 67–76 (2008)
Zor, C., Yanikoglu, B.A., Windeatt, T., Alpaydin, E.: FLIP-ECOC: a greedy optimization of the ECOC matrix. In: Proceedings of the 25th International Symposium on Computer and Information Sciences, London, UK, 22–24 September 2010, pp. 149–154 (2010)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Ismailoglu, F., Smirnov, E., Nikolaev, N., Peeters, R. (2015). Instance-Based Decompositions of Error Correcting Output Codes. In: Schwenker, F., Roli, F., Kittler, J. (eds) Multiple Classifier Systems. MCS 2015. Lecture Notes in Computer Science(), vol 9132. Springer, Cham. https://doi.org/10.1007/978-3-319-20248-8_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-20248-8_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-20247-1
Online ISBN: 978-3-319-20248-8
eBook Packages: Computer ScienceComputer Science (R0)