Abstract
Hybrid associative memories are based on the combination of two well-known associative networks, the lernmatrix and the linear associator, with the aim of taking advantage of their merits and overcoming their limitations. While these models have extensively been applied to information retrieval problems, they have not been properly studied in the framework of classification and even less with imbalanced data. Accordingly, this work intends to give a comprehensive response to some issues regarding imbalanced data classification: (i) Are the hybrid associative models suitable for dealing with this sort of data? and, (ii) Does the degree of imbalance affect the performance of these neural classifiers? Experiments on real-world data sets demonstrate that independently of the imbalance ratio, the hybrid associative memories perform poorly in terms of area under the ROC curve, but the hybrid associative classifier with translation appears to be the best solution when assessing the true positive rate.
Chapter PDF
Similar content being viewed by others
References
Anderson, J.A.: A simple neural network generating an interactive memory. Mathematical Biosciences 14(3-4), 197–220 (1972)
Deng, T., Zhao, L., Feng, L., Xue, W.: Information re-finding by context: A brain memory inspired approach. In: Proc. 20th ACM International Conference on Information and Knowledge Management, Glasgow, Scotland, UK, pp. 1553–1558 (2011)
Pantazi, S.V.: Methodological review: Unsupervised grammar induction and similarity retrieval in medical language processing using the Deterministic Dynamic Associative Memory (DDAM) model. Journal of Biomedical Informatics 43(5), 844–857 (2010)
Yaakobi, E., Bruck, J.: On the uncertainty of information retrieval in associative memories. In: Proc. IEEE International Symposium on Information Theory, Cambridge, MA, pp. 106–110 (2012)
Steinbuch, K.: Die Lernmatrix. Kybernetik 1, 36–45 (1961)
Anderson, J.A., Rosenfeld, E. (eds.): Neurocomputing: Foundations of Research. MIT Press, Cambridge (1988)
Kohonen, T.: Self-Organization and Associative Memory. Springer, New York (1989)
Santiago-Montero, R.: Clasificador Híbrido de Patrones basado en la Lernmatrix de Steinbuch y el Linear Associator de Anderson-Kohonen. Master’s thesis, Centro de Investigación en Computación, Instituto Politécnico Nacional, México D.F. (2003)
Japkowicz, N., Shaju, S.: The class imbalance problem: A systematic study. Intelligent Data Analysis 6(5), 429–449 (2002)
Galar, M., Fernández, A., Barrenechea, E., Bustince, H., Herrera, F.: A review on ensembles for the class imbalance problems: Bagging, boosting, and hybrid-based approaches. IEEE Trans. on Systems, Man, and Cybernetics – Part C: Applications and Reviews 42(4), 463–484 (2012)
García, V., Sánchez, J.S., Mollineda, R.A.: On the effectiveness of preprocessing methods when dealing with different levels of class imbalance. Knowledge-Based Systems 25(1), 13–21 (2012)
Akbani, R., Kwek, S., Japkowicz, N.: Applying support vector machines to imbalanced datasets. In: Proc. 15th European Conference on Machine Learning, Pisa, Italy, pp. 39–50 (2004)
Cieslak, D.A., Chawla, N.V.: Learning decision trees for unbalanced data. In: Proc. 19th European Conference on Machine Learning, Antwerp, Belgium, pp. 241–256 (2008)
Khoshgoftaar, T.M., Van Hulse, J., Napolitano, A.: Supervised neural network modeling: An empirical investigation into learning from imbalanced data with labeling errors. IEEE Trans. on Neural Networks 21(5), 813–830 (2010)
Hoens, T.R., Qian, Q., Chawla, N.V., Zhou, Z.-H.: Building decision trees for the multi-class imbalance problem. In: Tan, P.-N., Chawla, S., Ho, C.K., Bailey, J. (eds.) PAKDD 2012, Part I. LNCS, vol. 7301, pp. 122–134. Springer, Heidelberg (2012)
Cleofas-Sánchez, L., Guzmán Escobedo, M., Valdovinos, R.M., Yáñez Márquez, C., Camacho-Nieto, O.: Using hybrid associative classifier with traslation (HACT) for studying imbalance data sets. Engineering and Research 32(1), 53–57 (2012)
Hasson, M.H.: Associative Neural Memories: Theory and Implementation. Oxford University Press (1993)
Yáñez Márquez, C., Díaz de León, J.L.: Introducción a las Memorias Asociativas. Instituto Politécnico Nacional, México D.F. (2003)
Alcalá-Fernández, J., Fernández, A., Luengo, J., Derrac, J., García, S., Sánchez, L., Herrera, F.: KEEL data-mining software tool: Data set repository, integration of algorithms and experimental analysis framework. Journal of Multiple-Value Logic and Soft Computing 17(2-3), 255–287 (2011)
Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The WEKA data mining software: an update. SIGKDD Explorations Newsletter 11(1), 10–18 (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Cleofas-Sánchez, L., García, V., Martín-Félez, R., Valdovinos, R.M., Sánchez, J.S., Camacho-Nieto, O. (2013). Hybrid Associative Memories for Imbalanced Data Classification: An Experimental Study. In: Carrasco-Ochoa, J.A., Martínez-Trinidad, J.F., Rodríguez, J.S., di Baja, G.S. (eds) Pattern Recognition. MCPR 2013. Lecture Notes in Computer Science, vol 7914. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-38989-4_33
Download citation
DOI: https://doi.org/10.1007/978-3-642-38989-4_33
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-38988-7
Online ISBN: 978-3-642-38989-4
eBook Packages: Computer ScienceComputer Science (R0)