Abstract
Various methods for ensembles selection and classifier combination have been designed to optimize the performance of ensembles of classifiers. However, use of large number of features in training data can affect the classification performance of machine learning algorithms. The objective of this paper is to represent a novel feature elimination (FE) based ensembles learning method which is an extension to an existing machine learning environment. Here the standard 12 lead ECG signal recordings data have been used in order to diagnose arrhythmia by classifying it into normal and abnormal subjects. The advantage of the proposed approach is that it reduces the size of feature space by way of using various feature elimination methods. The decisions obtained from these methods have been coalesced to form a fused data. Thus the idea behind this work is to discover a reduced feature space so that a classifier built using this tiny data set would perform no worse than a classifier built from the original data set. Random subspace based ensembles classifier is used with PART tree as base classifier. The proposed approach has been implemented and evaluated on the UCI ECG signal data. Here, the classification performance has been evaluated using measures such as mean absolute error, root mean squared error, relative absolute error, F-measure, classification accuracy, receiver operating characteristics and area under curve. In this way, the proposed novel approach has provided an attractive performance in terms of overall classification accuracy of 91.11 % on unseen test data set. From this work, it is shown that this approach performs well on the ensembles size of 15 and 20.
Similar content being viewed by others
References
Acampora G, Lee CS, Vitiello A, Wang MH (2012) Evaluating cardiac health through semantic soft computing techniques. Soft Comput 16(7):1165–1181. doi:10.1007/s00500-011-0792-2
Blake CL, Merz CJ (1998) UCI repository of machine learning databases. http://www.ics.uci.edu/~mlearn/MLRepository.html
Breiman L (1996) Bagging predictors. Mach Learn 24:123–140
Breiman L (2001) Random forests. Mach Learn 45(1):5–32
Brown G (2009) Ensemble learning. Springer, New York
Ceylan R, Özbay Y (2007) Comparison of FCM, PCA and WT techniques for classification ECG arrhythmias using artificial neural network. Expert Syst Appl 33(2):286–295. doi:10.1016/j.eswa.2006.05.014
Chen SW (2000) A two stage discrimination of cardiac arrhythmia using a total least sqaure-based prony modeling algorithm. IEEE Trans Biomed Eng 47(10):1317–1327
Coast DA, Stern RM, Cano GG, Briller SA (1990) An approach to cardiac arrhythmia analysis using hidden markov models. IEEE Trans Biomed Eng 37(9):826–836
Cohen WW (1995) Fast effective rule induction. In: Proceedings of the 12th international conference on machine learning. Morgan Kaufmann, Lake Tahoe, California, pp 115–123
Dietterich TG (2000) Ensemble methods in machine learning. In: Multiple classifier system, pp 1–15
Dẑeroski S, Ẑenko B (2004) Is combining classifiers with stacking better than selecting the best one? Mach Learn 54(3):255–273 (special issue: meta-learning)
Elsayad AM (2009) Classification of ecg arrhythmia using learning vector quantization neural networks. In: Proceedings of the international conference on computer engineering and systems, ICCES 2009, pp 139–144
Exarchos TP, Tsipouras MG, Exarchos CP, Papaloukas C, Fotiadis DI, Michalis LK (2007) A methodology for the automated creation of fuzzy expert systems for ischaemic and arrhythmic beat classification based on a set of rules obtained by a decision tree. Artif Intell Med 40(3):187–200. doi:10.1016/j.artmed.2007.04.001
Frank E, Witten IH (1998) Generating accurate rule sets without global optimization. In: Proceeding of 15th international conference on machine learning, Morgan Kaufmann, San Francisco, pp. 144–151
Freund Y, Schapire RE (1996) Experiments with a new boosting algorithm. In: Thirteenth international conference on machine learning. Morgan Kaufmann, San Francisco, pp 148–156
Freund Y, Schapire RE (1997) A decision-theoretic generalization of on-line learning and an application to boosting. J Comput Syst Sci 55(1):119–139 Special issue for EuroCOLT ’95
García-Pedrajas N (2009) Constructing ensembles of classifiers by means of weighted instance selection. IEEE Trans Neu Netw 20(2):258–277. doi:10.1109/TNN.2008.2005496
Ge D, Srinivasan N, Krishnan S (2002) Cardiac arrhythmia classification using autoregressive modeling. BioMed Eng OnLine 1(1):5
Ge D-F, Hou B-P, Xiang X-J (2007) Study of feature based on auto regressive modeling in ecg automatic diagnosis. Acta Autom Sin 33(5):462–466. http://www.sciencedirect.com/science/article/pii/S1874102907600214
Gothwal H, Kedawat S, Kumar R (2011) Cardiac arrhythmias detection in an ecg beat signal using fast fourier transform and artificial neural network. J Biomed Sci Eng 4(04):289–296
Guyon I, Weston J, Barnhill S, Vapnik V (2002) Gene selection for cancer classification using support vector machines. Mach Learn 46:389–422
Hagan MT, Menhaj MB (1994) Training feedforward networks with the Marquardt algorithm. IEEE Trans Neural Netw 5(6):989–993
Hall MA (1998) Correlation-based feature subset selection for machine learning. Ph.D. thesis, University of Waikato, Hamilton
Hansen L, Salamon P (1990) Neural network ensembles. IEEE Trans Pattern Anal Mach Intel 12(10):993–1001
Harikumar R, Shivappriya S (2011) Analysis of qrs detection algorithm for cardiac abnormalities-a review. Intern J Soft Comput Eng 1(5):80–88
Haseena HH, Joseph PK, Mathew AT (2011) Classification of arrhythmia using hybrid networks. J. Medical System 35(6):1617–1630. doi:10.1007/s10916-010-9439-6
Haykin S (1999) Neural networks: a comprehensive introduction. Prentice Hall (1999)
Ho TK (1998) The random subspace method for constructing decision forests. IEEE Trans Pattern Anal Mach Intell 20(8):832–844
Hussain H, Fatt LL (2007) Efficient ecg signal classification using sparsely connected radial basis function neural network. In: Proceeding of the 6th WSEAS international conference on circuits, systems, electronics, control and, signal processing, Canada, pp 80–82
Issac Niwas S, Shantha Selva Kumari R, Sadasivam V (2005) Artificial neural network based automatic cardiac abnormalities classification. In: Proceeding of sixth international computational intelligence and multimedia applications conference, pp 41–46
Jadhav S, Nalbalwar S, Ghatol A (2011) Modular neural network model based foetal state classification. In: Proceeding of IEEE Interantional bioinformatics and biomedicine workshops (BIBMW) conference, pp 915–917
Jadhav SM, Nalbalwar SL, Ghatol A (2010) Artificial neural network based cardiac arrhythmia classification using ecg signal data. In: Proceeding of international electronics and information engineering (ICEIE) Conference. On, vol. 1, Kyoto
Jadhav SM, Nalbalwar SL, Ghatol AA (2010) Arrhythmia disease classification using artificial neural network model. In: Proceeding of IEEE international computational intelligence and computing research (ICCIC) conference, pp 1–4
Jadhav SM, Nalbalwar SL, Ghatol AA (2010) Ecg arrhythmia classification using modular neural network model. In: Proceeding of IEEE EMBS Conference biomedical engineering and sciences (IECBES), pp 62–66
Jadhav SM, Nalbalwar SL, Ghatol AA (2010) Generalized feedforward neural network based cardiac arrhythmia classification from ecg signal data. In: Proceeding of 6th international advanced information management and service (IMS) conference, pp 351–356
Jadhav SM, Nalbalwar SL, Ghatol AA (2011) Modular neural network based arrhythmia classification system using ECG signal data. Intern J Inf Technol Knowl Manag 4(I):205
Jadhav SM, Nalbalwar SL, Ghatol AA (2012) Artificial neural network models based cardiac arrhythmia disease diagnosis from ECG signal data (15/). http://research.ijcaonline.org/volume44/number15/pxc3878532.p
Kohavi R, John GH (1997) Wrappers for feature subset selection. Artif Intell 97(1—-2):273–324 Special issue on relevance
Kuncheva LI (2004) Combining pattern classifiers. Wiley, New York
Kuncheva LI, Rodriguez JJ, Plumpton CO, Linden DEJ, Johnston SJ (2010) Random subspace ensembles for fmri classification. IEEE Trans Med Imaging 29(2):531–542
Khadra Labib ASAF, Binajjaj S (2005) A quantitative analysis approach for cardiac arrhythmia classification using higher order spectral techniques. IEEE Trans Biomed Eng 52(11):1840–1845
Lagerholm M, Peterson C, Braccini G, Edenbrandt L, Sornmo L (2000) Clustering ecg complexes using hermite functions and self-organizing maps. IEEE Trans Biomed Eng 47(7):838–848
Lee SH, Uhm JK, Lim JS (2007) Extracting input features and fuzzy rules for detecting ecg arrhythmia based on newfm. In: Proceeding of international conference intelligent and advanced systems ICIAS 2007, pp 22–25
Mahmood AM, Kuppa MR (2012) A novel pruning approach using expert knowledge for data-specific pruning. Eng Comput (Lond) 28(1):21–30. doi:10.1007/s00366-011-0214-1
Moavenian M, Khorrami H (2010) A qualitative comparison of artificial neural networks and support vector machines in ECG arrhythmias classification. Expert Syst Appl 37(4):3088–3093. doi:10.1016/j.eswa.2009.09.021
Oveisi F, Oveisi S, Erfanian A, Patras I (2012) Tree-structured feature extraction using mutual information. IEEE Trans Neural Netw Learn Syst 23(1):127–137. doi:10.1109/TNNLS.2011.2178447
Owis MI, Abou-Zied AH, Youssef ABM, Kadah YM (2002) Study of features based on nonlinear dynamical modeling in ecg arrhythmia detection and classification. IEEE Trans Biomed Eng 49(7):733–736
Özbay Y, Ceylan R, Karlik B (2006) A fuzzy clustering neural network architecture for classification of ecg arrhythmias. Comput Biol Med 36(4):376–388
Perrone MP, Cooper LN (1993) When networks disagree: Ensemble methods for hybrid neural networks. In: Mammone RJ (ed.) Artificial neural networks for speech and vision, London, pp 126–142
Polat K, Sahan S, Günes S (2006) A new method to medical diagnosis: artificial immune recognition system (AIRS) with fuzzy weighted pre-processing and application to ECG arrhythmia. Expert Syst Appl 31(2):264–269
Quinlan R (1993) C4.5: Programs for machine learning. Morgan Kaufmann Publishers, San Mateo
Stiglic G, Kocbek S, Pernek I, Kokol P (2012) Comprehensive decision tree models in bioinformatics. PLoS One 7(3). http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=331
Ting KM, Witten IH (1997) Stacking bagged and dagged models. In: Proceeding of 14th international conference on machine learning. Morgan Kaufmann, pp 367–375
Uyar A, Gurgen F (2007) Arrhythmia classification using serial fusion of support vector machines and logistic regression. In: Proceeding of 4th IEEE workshop intelligent data acquisition and advanced computing systems: technology and applications IDAACS 2007, pp 560–565
Valentini G (2004) Random aggregated and bagged ensembles of SVMs: an empirical bias? variance analysis. In: Multiple classifier systems, pp 263–272
Valentini G, Dietterich TG (2004) Bias-variance analysis of support vector machines for the development of SVM-based ensemble methods. J Mach Learn Res 5:725–775
Wang Y, Zhu YS, Thakor NV, Xu YH (2001) A short-time multifractal approach for arrhythmia detection based on fuzzy neural, network. IEEE-Trans Biomed Eng 48:989–995
Waseem K, Javed A, Ramzan R, Farooq M (2011) Using evolutionary algorithms for ecg arrhythmia detection and classification. In: Natural computation (ICNC), 2011 seventh international conference on, vol. 4, pp 2386–2390
Yu SN, Chen YH (2007) Electrocardiogram beat classification based on wavelet transformation and probabilistic neural network. Pattern Recognit Lett 28(10):1142–1150
Zhao C, Gao Y, He J, Lian J (2012) Recognition of driving postures by multiwavelet transform and multilayer perceptron classifier. Eng Appl AI 25(8):1677–1686. doi:10.1016/j.engappai.2012.09.018
Zuo WM, Lu WG, Wang KQ, Zhang H (2008) Diagnosis of cardiac arrhythmia using kernel difference weighted knn classifier. In: Proceeding of computers in cardiology, pp 253–256
Acknowledgments
The authors would like to put on record their heart-felt thanks to the University Grants Commission (UGC), New Delhi and authority of Dr. Babasaheb Ambedkar Technological University, Lonere for providing ‘Teacher Fellowship Award’ to the corresponding author for his Ph.D. study. The corresponding author owe a sense of gratitude to Dr. B. B. Singh who helped in proof reading of this paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by G. Acampora.
Rights and permissions
About this article
Cite this article
Jadhav, S., Nalbalwar, S. & Ghatol, A. Feature elimination based random subspace ensembles learning for ECG arrhythmia diagnosis. Soft Comput 18, 579–587 (2014). https://doi.org/10.1007/s00500-013-1079-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-013-1079-6