Chaotic time series prediction using ELANFIS

This paper investigates the performance of Adaptive Neuro-Fuzzy Inference System (ANFIS) and Extreme Learning ANFIS (ELANFIS) in the chaotic time series prediction problem. ELANFIS is one of the neuro-fuzzy systems, which combines the learning capabilities of extreme learning machine (ELM) and the explicit knowledge of the fuzzy systems. In ELANFIS, premise parameters are randomly generated with some constraints to accommodate fuzziness, whereas consequent parameters are identified analytically using Moore-Penrose generalized inverse. Two benchmark problems, Mackey Glass equation and Lorenz equation, are used to compare the performance measures of the two algorithms. It has been shown that when the complexity of the model is increased the performance of ELANFIS is better than ANFIS because of the much lower training time required.

[1]  G. N. Pillai,et al.  Regularized extreme learning adaptive neuro-fuzzy algorithm for regression and classification , 2017, Knowl. Based Syst..

[2]  Zhiyuan Luo,et al.  Time series prediction with performance guarantee , 2011, IET Commun..

[3]  Chee Kheong Siew,et al.  Extreme learning machine: Theory and applications , 2006, Neurocomputing.

[4]  E. H. Mamdani,et al.  An Experiment in Linguistic Synthesis with a Fuzzy Logic Controller , 1999, Int. J. Man Mach. Stud..

[5]  Kumpati S. Narendra,et al.  Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.

[6]  K. V. Shihabudheen,et al.  Internal model control based on extreme learning ANFIS for nonlinear application , 2015, 2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems (SPICES).

[7]  Ravi Sankar,et al.  Time Series Prediction Using Support Vector Machines: A Survey , 2009, IEEE Computational Intelligence Magazine.

[8]  Li Tianliang,et al.  Prediction and analysis of chaotic time series on the basis of support vector , 2008 .

[9]  Michael Y. Hu,et al.  Forecasting with artificial neural networks: The state of the art , 1997 .

[10]  Jyh-Shing Roger Jang,et al.  ANFIS: adaptive-network-based fuzzy inference system , 1993, IEEE Trans. Syst. Man Cybern..

[11]  Dulakshi S. K. Karunasinghe,et al.  Chaotic time series prediction with a global model: Artificial neural network , 2006 .

[12]  A. Lapedes,et al.  Nonlinear Signal Processing Using Neural Networks , 1987 .

[13]  Chuen-Chien Lee,et al.  Fuzzy logic in control systems: fuzzy logic controller. II , 1990, IEEE Trans. Syst. Man Cybern..

[14]  Dianhui Wang,et al.  Extreme learning machines: a survey , 2011, Int. J. Mach. Learn. Cybern..

[15]  Pushpak Jagtap,et al.  Extreme learning ANFIS for control applications , 2014, 2014 IEEE Symposium on Computational Intelligence in Control and Automation (CICA).

[16]  Qinghua Zheng,et al.  Regularized Extreme Learning Machine , 2009, 2009 IEEE Symposium on Computational Intelligence and Data Mining.

[17]  Juan David Velasquez,et al.  Adaptive Multidimensional Neuro-Fuzzy Inference System for Time Series Prediction , 2015, IEEE Latin America Transactions.

[18]  Juan David Velasquez,et al.  Methodological Advances in Artificial Neural Networks for Time Series Forecasting , 2014, IEEE Latin America Transactions.

[19]  L. Glass,et al.  Oscillation and chaos in physiological control systems. , 1977, Science.

[20]  M. Sugeno,et al.  Structure identification of fuzzy model , 1988 .

[21]  Pushpak Jagtap,et al.  Comparison of extreme-ANFIS and ANFIS networks for regression problems , 2014, 2014 IEEE International Advance Computing Conference (IACC).

[22]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[23]  Hongming Zhou,et al.  Extreme Learning Machine for Regression and Multiclass Classification , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[24]  Alexey N. Averkin,et al.  Hybrid approach for time series forecasting based on ANFIS and Fuzzy Cognitive Maps , 2017, 2017 XX IEEE International Conference on Soft Computing and Measurements (SCM).