Abstract
The ARTMAP networks are machine learning techniques focused on supervised learning, being known mainly for their ability to learn fast, stable, incremental and online. Despite these advantages, the Fuzzy ARTMAP (FAM) suffers from the categories proliferation problem, leading to a reduction in its performance for unknown samples. Such disadvantage is mainly caused by the overlapping region (noise) between classes. The vast majority of work on this issue has been concerned with alleviating the problem. A technique used to improve the performance of a classifier is the rejection option, which is used to retain the classification of a sample if the decision is not considered to be reliable. Therefore, in this paper, we introduce a variant of the Fuzzy ARTMAP to behave as a classifier with the rejection option. The main idea is to create a region of rejection by looking at the place where the categories proliferate since it occurs precisely in the overlapping region. The proposal was validated by conducting experiments with real datasets, as well as by comparing them with other models (MLP, SVM, and SOM) applied with the same rejection option technique.
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
References
Grossberg, S.: Adaptive resonance theory: how a brain learns to consciously attend, learn, and recognize a changing world. Neural Netw. 37, 1–47 (2013)
Carpenter, G., et al.: Fuzzy ARTMAP: a neural network architecture for incremental supervised learning of analog multidimesional maps. IEEE Trans. Neural Netw. 3, 698–713 (1992)
El-Yaniv, R., Wiener, Y.: On the foundations of noise-free selective classification. J. Mach. Learn. Res. 11, 1605–1641 (2010)
Chow, C.: On optimum recognition error and reject tradeoff. IEEE Trans. Inf. Theory 16(1), 41–46 (1970)
Ishibuchi, H., Nii, M.: Neural networks for soft decision making. Fuzzy Sets Syst. 34(115), 121–140 (2000)
Sousa, R., da Rocha Neto, A.R., Barreto, G.A., Cardoso, J.S., Coimbra, M.T.: Reject option paradigm for the reduction of support vectors. In: ESANN (2014)
Marriott, S., Harrison, R.F.: A modified Fuzzy ARTMAP architecture for the approximation of noisy mappings. Neural Netw. 8(4), 619–641 (1995)
Xu, K.: How has the literature on Gini’s index evolved in the past 80 years? Working paper, Department of Economics, Dalhousie University (2003). https://doi.org/10.2139/ssrn.423200
Lichman, M.: UCI machine learning repository (2013)
Demšar, J.: Statistical comparisons of classifiers over multiple data sets. J. Mach. Learn. Res. 7, 1–30 (2006)
Sousa, R., Mora, B., Cardoso, J.S.: An ordinal data method for the classification with reject option. In: Proceedings of the International Conference on Machine Learning and Applications (ICMLA 2009), pp. 746–750 (2009)
Bounsiar, A., Beauseroy, P., Grall-Maes, E.: General solution and learning method for binary classification with performance constraints. Pattern Recogn. Lett. 29(10), 1455–1465 (2008)
Fumera, G., Roli, F.: Support vector machines with embedded reject option. In: Lee, S.-W., Verri, A. (eds.) SVM 2002. LNCS, vol. 2388, pp. 68–82. Springer, Heidelberg (2002). https://doi.org/10.1007/3-540-45665-1_6
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Sousa, F.F.M., Matias, A.L.S., da Rocha Neto, A.R. (2019). Classification with Rejection Option Using the Fuzzy ARTMAP Neural Network. In: Rojas, I., Joya, G., Catala, A. (eds) Advances in Computational Intelligence. IWANN 2019. Lecture Notes in Computer Science(), vol 11507. Springer, Cham. https://doi.org/10.1007/978-3-030-20518-8_47
Download citation
DOI: https://doi.org/10.1007/978-3-030-20518-8_47
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-20517-1
Online ISBN: 978-3-030-20518-8
eBook Packages: Computer ScienceComputer Science (R0)