Abstract
Neural network ensemble (NNE) is a simple and effective method to deal with incomplete data for classification. However, with the increase in the number of missing values, the number of incomplete feature combinations (feature subsets) grown rapidly which makes the NNE method very time-consuming and the accuracy is also need to be improved. In this paper, we propose a selective neural network ensemble (SNNE) classification for incomplete data. The SNNE first obtains all the available feature subsets of the incomplete dataset and then applies mutual information to measure the importance (relevance) degree of each feature subset. After that, an optimization process is applied to remove the feature subsets by satisfying the following condition: there is at least a feature subset contained in the removed feature subset and the difference of their importance degree is smaller than a given threshold δ. Finally, the rest of the feature subsets were used to train a group of neural networks and the classification for a given sample is decided by weighted majority voting of all available components in the ensemble. Experimental results show that δ = 0.05 is reasonable in our study. It can improve the efficiency of the algorithm without loss the algorithm accuracy. Experiments also show that SNNE outperforms the NNE-based algorithms compared. In addition, it can greatly reduce the running time when dealing with datasets with larger number of missing values.
Similar content being viewed by others
References
Allison PD (2001) Missing data, 136th edn. Sage publications, London
Little RJ, Rubin DB (2014) Statistical analysis with missing data. Wiley, New York
Lee SY, Song XY (2003) Maximum likelihood estimation and model comparison for mixtures of structural equation models with ignorable missing data. J Classif 20(2):221–255
Li X-L, Zhou Z-H (2007) Structure learning of probabilistic relational models from incomplete relational data. Machine Learning: ECML 2007. Springer, Berlin, pp 214–225
Duda RO, Hart PE (1973) Pattern classification and scene analysis, edn 3. Wiley, New York
Prati RC, Batista GE, Monard MC (2004) Class imbalances versus class overlapping: an analysis of a learning system behavior. MICAI 2004: advances in artificial intelligence. Springer, Berlin, pp 312–321
Rässler S (2004) The impact of multiple imputation for DACSEIS. University of Erlangen-Nurnberg, Germany
Williams D, Liao X-J, Xue Y, Carin L, Krishnapuram B (2007) On classification with incomplete data. IEEE Trans Pattern Anal Mach Intell 29(3):427–436
Clark PG, Grzymala B, Rzasa JW (2014) Mining incomplete data with singleton, subset and concept probabilistic approximations. Inf Sci 280:368–384
Ramoni M, Sebastianiv P (2001) Robust bayes classifiers. Artif Intell 125(1):209–226
Lin H-C, Su C-T (2013) A selective Bayes classifier with meta-heuristics for incomplete data. Neurocomputing 106:95–102
Ramoni M, Sebastiani P (2001) Robust learning with missing data. Mach Learn 45(2):147–170
Krause S, Polikar R (2003) An ensemble of classifiers approach for the missing feature problem. In: Proceedings of the 2003 international joint conference on neural networks, vol. 1, IEEE, pp 553–558
Jiang K, Chen H-X, Yuan S-M (2005) Classification for incomplete data using classifier ensembles. In: IEEE international conference on neural networks and brain, ICNN&B 2005, Vol 1, pp 559–563
Philippot E, Santosh KC, Belaïd A, Belaïd Y (2014) Bayesian networks for incomplete data analysis in form processing. Int J Mach Learn Cybernet 6(3):1–17
Wang X-Z, Wang R, Feng H-M, Wang H-C (2014) A new approach to classifier fusion based on upper integral. IEEE Trans Cybern 44(5):620–635
Wang X-Z, Xing H-J, Li Y, Hua Q, Dong C-R, Pedrycz W (2015) A study on relationship between generalization abilities and fuzziness of base classifiers in ensemble learning. IEEE Trans Fuzzy Syst 23(5):1638–1654
Chen H-X, Du Y-P, Jiang K (2012) Classification of incomplete data using classifier ensembles. In: 2012 IEEE international conference on systems and informatics, ICSAI2012, pp 2229–2232
Yan Y-T, Zhang Y-P, Zhang Y-W (2014) Multi-granulation ensemble classification for incomplete data. In: Rough sets and knowledge technology, Springer International Publishing, pp 343–351
Dai Q, Li M (2014) Introducing randomness into greedy ensemble pruning algorithms. Appl Intell 42(3):406–429
Zhang T, Dai Q, Ma Z (2015) Extreme learning machines’ ensemble selection with GRASP. Appl Intell 43(2):439–459
Ma Z, Dai Q, Liu N (2015) Several novel evaluation measures for rank-based ensemble pruning with applications to time series prediction. Exp Syst Appl 42(1):280–292
Wang X-Z, Chen A-X, Feng H-M (2011) Upper integral network with extreme learning mechanism. Neurocomputing 74(16):2520–2525
You Z-H, Lei Y-K, Zhu L et al (2013) Prediction of protein-protein interactions from amino acid sequences with ensemble extreme learning machines and principal component analysis. BMC Bioinformatics 14(Suppl 8):S10
Chen H, Ni D, Qin J et al (2015) Standard plane localization in fetal ultrasound via domain transferred deep neural networks. IEEE J Biomed Health Inform 19(5):1627–1636
Chen H-X, Yuan S-M, Jiang K (2005) Wrapper approach for learning neural network ensemble by feature selection. In: Advances in neural networks (ISNN 2005), vol 3496. Springer, Berlin, pp 526–531
Yu L, Liu H (2003) Feature selection for high-dimensional data: a fast correlation-based filter solution, In: ICML, Vol 3, pp 856–863
Erik S et al (2010) On estimating mutual information for feature selection, Artificial Neural Networks-ICANN 2010. Springer, Berlin, pp 362–367
Shannon CE (1948) A mathematical theory of communication. Bell Syst Tech J 27379–423:623–656
Benoit QF, Doquire G, Verleysen M (2014) Estimating mutual information for feature selection in the presence of label noise. Comput Stat Data Anal 71:832–848
Breiman L (1996) Bagging predictors. Mach Learn 24(2):123–140
UCI Repository of machine learning databases for classification, http://archive.ics.uci.edu/ml/datasets.html
Acknowledgments
This work was supported by National Natural Science Foundation of China (Nos. 61175046 and 61203290), Natural Science Foundation of Anhui Province (No. 1408085MF132).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Yan, YT., Zhang, YP., Zhang, YW. et al. A selective neural network ensemble classification for incomplete data. Int. J. Mach. Learn. & Cyber. 8, 1513–1524 (2017). https://doi.org/10.1007/s13042-016-0524-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13042-016-0524-0