Abstract
The prediction of unknown protein functions is one of the main concerns at field of computational biology. This fact is reflected specifically in the prediction of molecular functions such as catalytic and binding activities. This, along with the massive amount of information has made that tools based on machine learning techniques have increase their popularity in the last years. However, these tools are confronted to several problems associated to the treated data, one of them is the learning with large imbalance between their categories. There exist several techniques to overcomes the class imbalance, but most of them present many weakness that difficult the obtaining of reliable results. Moreover, models based on cost sensitive learning seems to be a good choice to deal with imbalance data, yet, the obtaining of a optimal cost matrix still remains an open issue. In this paper, a methodology to calculate a optimal cost matrix for models based on cost sensitive learning is proposed. The results show the superiority of this approach compared with several techniques in the state of the art regarding to class imbalance. Tests were applied to prediction of molecular functions in Embryophyta plants.
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 subscriptionsReferences
Aebersold, R., Mann, M., et al.: Mass spectrometry-based proteomics. Nat. 422(6928), 198–207 (2003)
Allison, D.B., Cui, X., Page, G.P., Sabripour, M.: Microarray data analysis: from disarray to consolidation and consensus. Nat. Rev. Genet. 7(1), 55–65 (2006)
Ashburner, M., Ball, C.A., Blake, J.A., Botstein, D., Butler, H., Cherry, J.M., Davis, A.P., Dolinski, K., Dwight, S.S., Eppig, J.T., et al.: Gene ontology: tool for the unification of biology. Nat. Genet. 25(1), 25 (2000)
Basu, M.: Data Complexity in Pattern Recognition. Springer, New York (2006)
Chawla, N.V., Bowyer, K.W., Hall, L.O., Kegelmeyer, W.P.: Smote: synthetic minority over-sampling technique. J. Artif. Intell. Res. 16, 321–357 (2002)
Dietterich, T.G.: An experimental comparison of three methods for constructing ensembles of decision trees: bagging, boosting, and randomization. Mach. Learn. 40(2), 139–157 (2000)
Ding, Z.: Diversified ensemble classifiers for highly imbalanced data learning and its application in bioinformatics. Ph.D thesis, Georgia State University (2011)
Domingos, P.: Metacost: a general method for making classifiers cost-sensitive. In: Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 155–164. ACM (1999)
Elkan, C.: The foundations of cost-sensitive learning. In: International Joint Conference on Artificial Intelligence, vol. 17, pp. 973–978. Lawrence Erlbaum Associates Ltd (2001)
García-López, S., Jaramillo-Garzón, J.A., Higuita-Vásquez, J.C., Castellanos-Domínguez, C.G.: Wrapper and filter metrics for PSO-based class balance applied to protein subcellular localization. In: 2012 Biostec-Bioinformatics (2012)
Grzymala-Busse, J.W., Stefanowski, J., Wilk, S.: A comparison of two approaches to data mining from imbalanced data. J. Intell. Manuf. 16(6), 565–573 (2005)
He, H., Garcia, E.A.: Learning from imbalanced data. IEEE Trans. Knowl. Data Eng. 21(9), 1263–1284 (2009)
Ho, T.K., Basu, M.: Complexity measures of supervised classification problems. IEEE Trans. Pattern Anal. Mach. Intell. 24(3), 289–300 (2002)
Jain, E., Bairoch, A., Duvaud, S., Phan, I., Redaschi, N., Suzek, B., Martin, M., McGarvey, P., Gasteiger, E.: Infrastructure for the life sciences: design and implementation of the uniprot website. BMC Bioinform. 10(1), 136 (2009)
Jaramillo-Garzón, J.A., Gallardo-Chacón, J.J., Castellanos-Domínguez, C.G., Perera-Lluna, A.: Predictability of gene ontology slim-terms from primary structure information in embryophyta plant proteins. BMC Bioinform. 14(1), 68 (2013)
Larrañaga, P., Calvo, B., Santana, R., Bielza, C., Galdiano, J., Inza, I., Lozano, J.A., Armañanzas, R., Santafé, G., Pérez, A., et al.: Machine learning in bioinformatics. Briefings Bioinform. 7(1), 86–112 (2006)
Liu, X.Y., Zhou, Z.H.: The influence of class imbalance on cost-sensitive learning: an empirical study. In: 2006 Sixth International Conference on Data Mining, ICDM’06, pp. 970–974. IEEE (2006)
Liu, X.-Y., Zhou, Z.-H.: Towards cost-sensitive learning for real-world applications. In: Cao, L., Huang, J.Z., Bailey, J., Koh, Y.S., Luo, J. (eds.) PAKDD Workshops 2011. LNCS, vol. 7104, pp. 494–505. Springer, Heidelberg (2012)
Polikar, R.: Ensemble based systems in decision making. IEEE Circuits Syst. Mag. 6(3), 21–45 (2006)
Schapire, R.E.: A brief introduction to boosting. In: International Joint Conference on Artificial Intelligence, vol. 16, pp. 1401–1406. Lawrence Erlbaum Associates Ltd (1999)
Sonnenburg, S., Schweikert, G., Philips, P., Behr, J., Rätsch, G.: Accurate splice site prediction using support vector machines. BMC Bioinform. 8(Suppl 10), S7 (2007)
Su, C.T., Hsiao, Y.H.: An evaluation of the robustness of MTS for imbalanced data. IEEE Trans. Knowl. Data Eng. 19(10), 1321–1332 (2007)
Mohanna, E., Valian, E., Tavakoli, S.: Improved cuckoo search algorithm for global optimization. Int. J. Commun. Inf. Technol. 1(1), 31–44 (2011)
Yang, P., Xu, L., Zhou, B.B., Zhang, Z., Zomaya, A.Y.: A particle swarm based hybrid system for imbalanced medical data sampling. BMC Genomics 10(Suppl 3), S34 (2009)
Acknowledgements
This work is partially funded by the Research office (DIMA) at the Universidad Nacional de Colombia at Manizales and the Colombian National Research Centre (COLCIENCIAS) through grant No.111952128388 and the “jovenes investigadores e innovadores - 2010 Virginia Gutierrez de Pineda” fellowship.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
García-López, S., Jaramillo-Garzón, J.A., Castellanos-Dominguez, G. (2014). Optimization of Cost Sensitive Models to Improve Prediction of Molecular Functions. In: Fernández-Chimeno, M., et al. Biomedical Engineering Systems and Technologies. BIOSTEC 2013. Communications in Computer and Information Science, vol 452. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-662-44485-6_15
Download citation
DOI: https://doi.org/10.1007/978-3-662-44485-6_15
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-662-44484-9
Online ISBN: 978-3-662-44485-6
eBook Packages: Computer ScienceComputer Science (R0)