Abstract
The selection of the optimal ensembles of classifiers in multiple-classifier selection technique is un-decidable in many cases and it is potentially subjected to a trial-and-error search. This paper introduces a quantitative meta-learning approach based on neural network and rough set theory in the selection of the best predictive model. This approach depends directly on the characteristic, meta-features of the input data sets. The employed meta-features are the degree of discreteness and the distribution of the features in the input data set, the fuzziness of these features related to the target class labels and finally the correlation and covariance between the different features. The experimental work that consider these criteria are applied on twenty nine data sets using different classification techniques including support vector machine, decision tables and Bayesian believe model. The measures of these criteria and the best result classification technique are used to build a meta data set. The role of the neural network is to perform a black-box prediction of the optimal, best fitting, classification technique. The role of the rough set theory is the generation of the decision rules that controls this prediction approach. Finally, formal concept analysis is applied for the visualization of the generated rules.
Similar content being viewed by others
References
Gibert K, Sanchez-Marre M, Codina V (2010) Choosing the right data mining technique: classification of methods and intelligent recommendation. In: Proceedings of the IEMSs fifth biennial meeting: international congress on, environmental modelling and software, pp 1933–1940
Zhuang J, Widschwendter M, Teschendorff AE (2012) A comparison of feature selection and classification methods in DNA methylation studies using the Illumina Infinium platform. BMC Bioinforma 13(1):59–74
Sharkey AJC, Sharkey NE (1997) Combining diverse neural nets. Knowl Eng Rev 12(3):231–247
Ruta D, Gabrys B (2005) Classifier selection for majority voting. Inf Fusion 6:63–81
Dimililer N, Varolu E, Altnay H (2007) Vote-based classifier selection for biomedical NER using genetic algorithms. LNCS 4478:202–209
Matti A (2003) Comparison of classifier selection methods for improving committee performance. In: Proceedings of the 4th international conference on multiple classifier systems. Guildford, pp 84–93
Juran JM, Blanton Godfrey A (1999) Juran’s quality handbook, 5th edn. McGraw-Hill, New York
Aho T, Elomaa T, Kujala J (2008) Unsupervised classifier selection based on two-sample test. In: Proceedings of the 11th international conference on discovery science, Budapest, pp 2839
Salama MA, Hassanien AE, Fahmy AA (2010) Pattern-based subspace classification model. In: The second world congress on nature and biologically inspired computing (NaBIC2010), Kitakyushu, Japan, pp 357–362
Phyu TN (2009) Survey of classification techniques in data mining. In: Proceedings of the international multiconference of engineers and computer scientists (IMECS). Hong Kong, vol 1, pp 727–731
Geurts P (2001) Pattern extraction for time series classification. In: Proceedings of the 5th European conference on principles of data mining and knowledge discovery, Freiburg, Germany, pp 115–127
Vilalta R, Giraud-Carrier C, Brazdil P, Soares C (2004) Using meta-learning to support data mining. Int J Comput Sci Appl 1:31–45
Brazdil P, Giraud-Carrier C, Soares C, Vilalta R (2009) Metalearning: applications to data mining. Springer, Berlin
Prudêncio Ricardo BC, de Souto Marcilio CP, Ludermir TB (2011) Selecting machine learning algorithms using the ranking meta-learning approach. Stud Comput Intell 358:225–243
Prudêncio Ricardo BC, Soares C, Ludermir TB (2011) Combining meta-learning and active selection of datasetoids for algorithm selection. Lect Notes Comput Sci 6678:164–171
Ferrari DG, de Castro LN (2012) Clustering algorithm recommendation: a meta-learning approach, swarm, evolutionary, and memetic computing. Lect Notes Comput Sci 7677:143–150
Villar JR, González S, Sedano J, Corchado E (2012) Meta-heuristic improvements applied for steel sheet incremental cold shaping. Memetic Comput
Figueredo GP, Ebecken NFF, Augusto DA, Barbosa HJC (2012) An immune-inspired instance selection mechanism for supervised classification. Memetic Comput 4(2):135–147
Priya R (2011) Predicting execution time of machine learning tasks using metalearning, Information and Communication Technologies (WICT). World Congress on Dec. 2011, pp 1193–1198
Maszczyk T, Grochowski M, Duch W (2010) Discovering data structures using meta-learning, visualization and constructive neural networks. Stud Comput Intell 262:467–484
Cruz-Reyes L, Gómez-Santillán C, Pérez-Ortega J, Landero V, Quiroz M, Ochoa A (2012) Algorithm selection: from meta-learning to hyper-heuristics, intelligent systems, ISBN: 978-953-51-0054-6, InTech, doi:10.5772/36710. Available from: http://www.intechopen.com/books/intelligent-systems/algorithm-selection-from-meta-learning-to-hyper-heuristics
Salama MA, Revett K, Hassanien AE, Fahmy AA (2011) Interval-based attribute evaluation algorithm. In: The 6th IEEE international symposium advances in artificial intelligence and applications, Szczecin, Poland, Sep 18–21, pp 153–156
Phyu TN (2009) Survey of classification techniques in data mining. In: Proceedings of the international multiconference of engineers and computer scientists, IMECS 2009. Hong Kong, vol 1, pp 727–731
Kaytoue M, Duplessis S, Kuznetsov SO, Napoli A (2009) Two FCA-based methods for mining gen expression data. Lect Notes Comput Sci 5548:251–266
Mastrogiannis N, Boutsinas B, Giannikos I (2009) A method for improving the accuracy of data mining classification algorithms. Comput Oper Res 36(10):2829–2839
Geurts P (2001) Pattern extraction for time series classification. In: Proceedings of the 5th European conference on principles of, data mining and knowledge discovery, pp 115–127
O’Rourke N, Hatcher L, Stepanski EJ (2005) A step-by-step approach to using SAS for univariate and multivariate statistics, 2nd edn. SAS Institute Inc, USA. ISBN 1-59047-417-1
Rosenbaum PR (2010) Causal inference in randomized experiments. Springer Ser Stat Design Observ Stud 1:21–63
Salama MA, Hassanien AE, Fahmy AA (2010) Reducing the influence of normalization on data classification. In: The 6th international conference on next generation web services practices (NWeSP 2010), Gwalior, India, pp 609–703
Carmen L, Reinders MJT, Wessels LFA (2006) Random subspace method for multivariate feature selection. Pattern Recogn Lett 10:1067–1076
Shang C, Shen Q (2006) Aiding classification of gene expression data with feature selection: a comparative study. Comput Intell Res 1:68–76
Liu H, Yu L (2005) Toward integrating feature selection algorithms for classification and clustering. IEEE Trans Knowl Data Eng 17:1–12
Doak J (1992) An evaluation of feature selection methods and their application to computer security (Tech. Rep. CSE-92-18). University of California at Davis
Pawlak Z (1997) Rough set approach to knowledge-based decision support. Eur J Oper Res 99:48–57
Mak B, Munakata T (2002) Rule extraction from expert heuristics: a comparative study of rough sets with neural networks and ID3. Eur J Oper Res 136(1):212–229
Nguyen HS, Nguyen SH (1998) Discretization methods in data mining. In: Polkowski L, Skowron A (eds) Rough sets in knowledge discovery, vol 1. Physica-Verlag, pp 451–482
Salama AS (2011) Some topological properties of rough sets with tools for data mining. IJCSI Int J Comput Sci, Issues 8(3), No. 2
Cabestany J, Prieto A, Sandoval DF (2005) Heuristic search over a ranking for feature selection. LNCS 3512:742–749
Marsaglia G, Tsang WW, Wang J (2003) Evaluating Kolmogorovs distribution. J Stat Softw 8(18):1–4
Varki S, Cooil B, Rust RT (2000) Modelling fuzzy data in qualitative market research. J Market Res 37(4):480–489
Yang L (2005) Uniformization of Discrete Data. LNCS 3827:453–462
Maglogiannis I, Loukis E, Zafiropoulos E, Stasis A (2009) Support vectors machine-based identification of heart valve diseases using heart sounds. J Comput Methods Progr Biomed 95:47–61
Chiba University hospital DataBase. http://lisp.vse.cz/pkdd99/
http://kdd.ics.uci.edu/databases. Irvine, CA, USA, July, 2010
UCI Machine Learning Repository. http://archive.ics.uci.edu/ml/datasets.html
Weka: Data Mining Software in java. http://www.cs.waikato.ac.nz/ml/weka/
Chao S, Li Y (2005) Multivariate interdependent discretization for continuous attribute. In: Proceeding of the 3rd international conference on information technology and applications, vol 1, pp 167–172
Salama MA, Hassanien AE (2012) Binarization and validation in formal concept analysis. Int J Syst Biol Biomed Technol 1(4): 17–28
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Salama, M.A., Hassanien, A.E. & Revett, K. Employment of neural network and rough set in meta-learning. Memetic Comp. 5, 165–177 (2013). https://doi.org/10.1007/s12293-013-0114-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12293-013-0114-6