Abstract
Computational complexity and sample selection are two main factors that limited the performance of online sequential extreme learning machines (OS-ELMs). This paper proposes a new model that introduces the concept of hybrid kernel and sample selection method based on an online learning model using a membership function. In other words, an online sequential extreme learning machine based on a hybrid kernel function (HKOS-ELM) is presented. The algorithm only calculates the kernel function to determine the final output function, mostly solving the computational complexity of the algorithm. The hybrid kernel function proposed in this paper has the advantages of strong learning ability and good generalization performance of single kernel function. Based on the classification essence of the OS-ELM classification, the membership function is introduced into the sample selection to remove the noise point and the outlier point. The experimental results showed that the HKOS-ELM algorithm adding the membership degree with mixed kernel functions preserves the advantages of kernel functions and online learning and improves the classification performance of the system.







Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.References
Huang GB, Zhu QY, Siew CK (2004) Extreme learning machine: a new learning scheme of feedforward neural network. In: Proceedings of international joint conference on neural networks (IJCNN’04), 2004, pp 985–501
Huang GB, Zhu QY, Siew CK (2006) Extreme learning machine: theory and applications. Neural Comput 70:489–501
Liang NY, Huang GB (2006) A fast accurate online sequential learning algorithm for feed forward networks. IEEE Trans Neural Netw 17:1411–1423
Le Y, Lei Y (2013) Online sequence ELM model based on the kernel function. Basic Sci J Text Univ 26(4):516–520
Zhang G, Xie XS, Huang Y, Wang C (2014) An online multi-kernel learning algorithm for big data. CAAI Trans Intell Syst 9(3):355–363
Wang JG, Wu LM, Zhang WX, Jiang X (2014) Online support vector machine based on hybrid kernel function and its application in methanol synthesis. Mach Des Manuf (8):217–219
Simone S, Danilo C, Michele S, Aurelio U (2015) Online sequential extreme learning machine with kernels. IEEE Trans Neural Netw Learn Syst 26(9):2214–2220
Xu Y-G, Deng W-K, Chen L-D (2014) Online fault diagnosis based on kernel-based weighted extreme learning machine in wastewater treatment process. [EB/OL]. http://www.cnki.net/kcms/detail/11.1946.TQ.20160511.1522.016.html. Accessed 25 Aug 2014
Chao MA, Zhang YT, Li ZN, Yin G (2014) Hydraulic pump characteristic parameters online prediction based on kernel extreme learning machine. Comput Simul 31(5):351–354
Ding SF, Zhang YN, Xu XZ, Bao L (2013) A novel extreme learning machine based on hybrid kernel function. JCP 8(8):2110–2116
Mo YB, Xu SH (2010) Application of SVM based on hybrid kernel function in heart disease diagnoses. In: 2010 International conference on intelligent computing and cognitive informatics
Cai DL, Guo H (2014) Protein-protein interaction prediction method based on SVM with hybrid kernel function. J Fuzhou Univ (Nat Sci Edn) 42(6):006
Liu CW, Luo JX (2014) A PSO-SVM classifier bases on hybrid kernel function. J East China Univ Sci Technol (Nat Sci Edn) 40(1):96–101
Song W, Yi T, Xin Z, Xin Z (2014) The research on hybrid kernel function facing vehicle driving pattern recognition. In: Proceedings of 2014 international conference on industrial engineering and information technology
Liu M, Zhou SS, Wu H (2009) SHM based on new hybrid kernel function. J Comput Appl 29(S2):167–168+206
Tang Q, Wang HR, Xu XY, Wang C (2014) Hydrological time series model based on SVM with hybrid kernel function and its application. Syst Eng Theory Pract 34(2):521–529
Yan GT, Ma GF, Xiao YZ (2007) Support vector machines based on hybrid kernel function. J Harbin Inst Technol 39(11):1704–1706
Wei JR (2015) A reevaluation of hybrid kernel function for support vector machine. Stat Res 32(2):90–96
Qing M (2013) Instance selection research based on ELM and PNN. Hebei University, Hebei, p 6
Mo C (2014) Network traffic prediction model based on selecting training samples with reasonable forgetting. Comput Appl Softw 31(10):120–123
Du ZL, Li XM, Zheng ZG, Zhang GR, Mao Q (2015) Extreme learning machine based on regularization and forgetting factor and its application in fault prediction. Chin J Sci Instrum 36(7):1546–1553
Zhang X, Wang H-L (2011) Selective forgetting extreme learning machine and its application to time series prediction. Acta Phys Sin 60(8):080504
Zhang Y, Zhu X, Luo Y (2014) An SVM Algorithm for overcoming the influence of muscle fatigue in sEMG based human machine interaction. Control Eng China 21(4):96–101
Wu C-M, Wang X-D, Bai D-Y, Zhang H-D (2010) Fast SVM incremental learning algorithm using KKT conditions and between-class convex hull vectors. Comput Eng Des 31(8):1792–1798
Xiao ZC, Wang J, Wang Y-S (2010) Predict the time series of the parameter-varying chaotic system based on recursive lease square support vector machine (RLS-SVM). Aeronaut Comput Tech 40(3):30–37
Lu YL, Li L, Zhou MM, Tian G-L (2009) A new fuzzy support vector machine based on hybrid kernel function. In: Proceedings of the eighth international conference on machine learning and cybernetics, Baoding, 12–15 July 2009
Rong HJ, Huang GB (2009) Online sequential fuzzy extreme learning machine for function approximation and classification problems. IEEE Trans Syst Man Cybern Part B Cybern 39(4):1067–1072
Dhillon IS, Modha DS (2001) Concept decomposition for large sparse text data using clustering. Mach Learn 42(1):143–175
Chiang JH, Hao PY (2003) A new kernel-based fuzzy clustering approach: support vector clustering with cell growing. IEEE Trans Fuzzy Syst 11(4):518–527
Zhao K, Li L, Deng N (2012) A new method to construct fuzzy membership. Comput Technol Dev 22(8):75–77
Lin CF, Wang SD (2002) Fuzzy support vector machines. IEEE Trans Neural Netw 13(2):464–471
Liu SY, Du Z (2007) An improved fuzzy support vector machine method. CAAI Trans Intell Syst 2(3):30–33
Ye Y, Squartini S, Piazza F (2012) Online sequential extreme learning machine in no stationary environments. Neurocomputing 116:94–101
Zhang S, Liu J, Tian J-W (2004) An SVM-based small target segmentation and clustering approach. In: Proceedings of the third international conference on machine learning and cybernetics. IEEE, Shanghai, pp 3318–3323
Wille R (1982) Restructuring lattice theory: an approach based on hierarchies of concepts. In: Rival I (ed) Ordered Sets. Reidel, Dordrecht, pp 445–470
Christopher B (2007) Pattern recognition and machine learning. Springer, New York
Du Z (2009) Research on some variants of support vector machine. Xi Dian University, Xi’an
Acknowledgements
The paper is supported in part by the National Natural Science Foundation of China (NSFC) under Grant Nos. 61272036, 61702213.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Zhang, S., Tan, W., Wang, Q. et al. A new method of online extreme learning machine based on hybrid kernel function. Neural Comput & Applic 31, 4629–4638 (2019). https://doi.org/10.1007/s00521-018-3629-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-018-3629-4