Advertisement

Time Series Prediction Using Complex-Valued Legendre Neural Network with Different Activation Functions

  • Bin Yang
  • Wei Zhang
  • Haifeng Wang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10956)

Abstract

In order to enchance the flexibility and functionality of Legendre neural network (LNN) model, complex-valued Legendre neural network (CVLNN) is proposed to predict time series data. Bat algorithm is proposed to optimize the real-valued and complex-valued parameters of CVLNN model. We investigate performance of CVLNN for predicting small-time scale traffic measurements data by using different complex-valued activation functions like Elliot function, Gaussian function, Sigmoid function and Secant function. Results reveal that Elliot function and Sigmoid function predict more accurately and have faster convergence than Gaussian function and Secant function.

Keywords

Complex-valued Legendre neural network Activation function Bat algorithm 

Notes

Acknowledgments

This work was supported by the Natural Science Foundation of China (No. 61702445), the PhD research startup foundation of Zaozhuang University (No. 2014BS13), Zaozhuang University Foundation (No. 2015YY02), and Shandong Provincial Natural Science Foundation, China (No. ZR2015PF007).

References

  1. 1.
    Angeline, P.J., Saunders, G.M., Pollack, J.B.: An evolutionary algorithm that constructs recurrent neural networks. IEEE Trans. Neural Networks 5, 54–65 (1994)CrossRefGoogle Scholar
  2. 2.
    Patra, J.C., Meher, P.K., Chakraborty, G.: Nonlinear channel equalization for wireless communication systems using Legendre neural networks. Sig. Process. 89(11), 2251–2262 (2009)CrossRefGoogle Scholar
  3. 3.
    Pei, A.Q., Wang, J., Fang, W.: Predicting agent-based financial time series model on lattice fractal with random Legendre neural network. Soft Comput., 1–16 (2015)Google Scholar
  4. 4.
    Dash, R., Dash, P.K.: Prediction of financial time series data using hybrid evolutionary legendre neural network. Int. J. Appl. Evol. Comput. 7(1), 16–32 (2016)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Behera, S., Sahu, B.: Non linear dynamic system identification using Legendre neural network and firefly algorithm. In: 2016 International Conference on Communication and Signal Processing (ICCSP), pp. 1689–1693. IEEE, MADRAS, India (2016)Google Scholar
  6. 6.
    Wang, C., Qin, S.Y., Wan, B.W.: A novel neural network structure with fast convergence based on optimizing combination of different activation function. In: International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 1399–1400. IEEE, Orlando, FL, USA (1991)Google Scholar
  7. 7.
    Zhang, L.M.: Genetic deep neural networks using different activation functions for financial data mining. In: IEEE International Conference on Big Data, pp. 2849–2851. IEEE, Santa Clara, CA, USA (2015)Google Scholar
  8. 8.
    Kumar, A.J.P., Singh, D.K.J.: A study on different activation functions for neural network-based wear loss prediction. Int. J. Appl. Eng. Res. 4(3), 399 (2009)MathSciNetGoogle Scholar
  9. 9.
    Malleswaran, M., Dr, V.V., Angel, D.S.: Data fusion using different activation functions in artificial neural networks for vehicular navigation. Int. J. Eng. Sci. Technol. 2(12), 7676–7690 (2010)Google Scholar
  10. 10.
    Yang, X.S., He, X.S.: Bat algorithm: literature review and applications. Int. J. Bio-Inspired Comput. 5(3), 141–149 (2013)CrossRefGoogle Scholar
  11. 11.
    Rahmani, M., Ghanbari, A., Ettefagh, M.M.: Robust adaptive control of a bio-inspired robot manipulator using bat algorithm. Expert Syst. Appl. 56(C), 164–176 (2016)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.School of Information Science and EngineeringZaozhuang UniversityZaozhuangChina

Personalised recommendations