Advertisement

Neural Computing and Applications

, Volume 31, Issue 1, pp 307–316 | Cite as

Median-Pi artificial neural network for forecasting

  • Erol Egrioglu
  • Ufuk YolcuEmail author
  • Eren Bas
  • Ali Zafer Dalar
Original Article
  • 207 Downloads

Abstract

Datasets with outliers can be predicted with robust learning methods or robust artificial neural networks. In robust artificial neural networks, the architectures become robust by using robust statistics as aggregation functions. Median neural network and trimmed mean neural network are two robust artificial neural networks used in the literature. In these robust artificial neural networks, median and trimmed mean statistics are used as aggregation functions. In this study, Median-Pi artificial neural network is proposed as a new robust neural network for the purpose of forecasting. In Median-Pi artificial neural network, median and multiplicative functions are used as aggregation functions. Because of using median, the proposed network can produce good results for data with outliers. The Median-Pi artificial neural network is trained by particle swarm optimization. The performance of the neural network is investigated by using datasets from the International Time Series Forecast Competition 2016 (CIF-2016). The performance of the proposed method in case of outlier is compared to some other artificial neural networks. Median neural network, trimmed mean neural network, Pi-Sigma neural network and the proposed robust network are applied to time series with outlier, and the obtained results are compared. According to application results, the proposed Median-Pi artificial neural network can produce better forecast results than the other network types.

Keywords

Forecasting Outliers Particle swarm optimization Robust artificial neural networks 

Notes

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Supplementary material

521_2017_3002_MOESM1_ESM.docx (19 kb)
ESM 1 (DOCX 19 kb)
521_2017_3002_MOESM2_ESM.docx (19 kb)
ESM 2 (DOCX 19 kb)
521_2017_3002_MOESM3_ESM.docx (20 kb)
ESM 3 (DOCX 19 kb)
521_2017_3002_MOESM4_ESM.docx (19 kb)
ESM 4 (DOCX 19 kb)
521_2017_3002_MOESM5_ESM.docx (19 kb)
ESM 5 (DOCX 19 kb)
521_2017_3002_MOESM6_ESM.docx (19 kb)
ESM 6 (DOCX 19 kb)
521_2017_3002_MOESM7_ESM.docx (19 kb)
ESM 7 (DOCX 19 kb)
521_2017_3002_MOESM8_ESM.docx (19 kb)
ESM 8 (DOCX 19 kb)
521_2017_3002_MOESM9_ESM.docx (19 kb)
ESM 9 (DOCX 19 kb)
521_2017_3002_MOESM10_ESM.docx (19 kb)
ESM 10 (DOCX 19 kb)
521_2017_3002_MOESM11_ESM.docx (19 kb)
ESM 11 (DOCX 19 kb)
521_2017_3002_MOESM12_ESM.docx (19 kb)
ESM 12 (DOCX 19 kb)
521_2017_3002_MOESM13_ESM.docx (19 kb)
ESM 13 (DOCX 19 kb)
521_2017_3002_MOESM14_ESM.docx (19 kb)
ESM 14 (DOCX 19 kb)
521_2017_3002_MOESM15_ESM.docx (19 kb)
ESM 15 (DOCX 19 kb)
521_2017_3002_MOESM16_ESM.docx (19 kb)
ESM 16 (DOCX 19 kb)
521_2017_3002_MOESM17_ESM.docx (19 kb)
ESM 17 (DOCX 19 kb)
521_2017_3002_MOESM18_ESM.docx (19 kb)
ESM 18 (DOCX 19 kb)
521_2017_3002_MOESM19_ESM.docx (19 kb)
ESM 19 (DOCX 19 kb)
521_2017_3002_MOESM20_ESM.docx (19 kb)
ESM 20 (DOCX 19 kb)

References

  1. 1.
    Chen DS, Jain RC (1994) A robust backpropagation learning algorithm for function approximation. IEEE Trans Neural Netw 5:467–479. doi: 10.1109/72.286917 CrossRefGoogle Scholar
  2. 2.
    Hsiao C-C, Chuang C-C, Jeng J-T (2012) Robust back propagation learning algorithm based on near sets. In: 2012 Int. Conf. Syst. Sci. Eng. (ICSSE),June 30–July 2, 2012, Dalian, China. pp 19–23Google Scholar
  3. 3.
    Lee CC, Chung PC, Tsai JR, Chang CI (1999) Robust radial basis function neural networks. IEEE Trans Syst Man, Cybern Part B Cybern 29:674–685. doi: 10.1109/3477.809023 CrossRefGoogle Scholar
  4. 4.
    El-Melegy MT, Essai MH, Ali AA (2009) Robust training of artificial feedforward neural networks. In: Found. Comput. Intell. Vol. 1 Vol. 201 Ser. Stud. Comput. Intell. pp 217–242Google Scholar
  5. 5.
    Rusiecki A (2012) Robust learning algorithm based on iterative least median of squares. Neural Process Lett 36:145–160CrossRefGoogle Scholar
  6. 6.
    Thomas P, Bloch G, Sirou F, Eustache V (1999) Neural modeling of an induction furnace using robust learning criteria. (I. Press, Éd.) Integrated Computer-Aided Engineering 6(1):15–26CrossRefGoogle Scholar
  7. 7.
    Bors AG, Pitas I (1996) Median radial basis function neural network. IEEE Trans Neural Netw. doi: 10.1109/72.548164 Google Scholar
  8. 8.
    Majhi B, Rout M, Majhi R et al (2012) New robust forecasting models for exchange rates prediction. Expert Syst Appl 39:12658–12670. doi: 10.1016/j.eswa.2012.05.017 CrossRefGoogle Scholar
  9. 9.
    Aladag CH, Egrioglu E, Yolcu U (2014) Robust multilayer neural network based on median neuron model. Neural Comput Appl. doi: 10.1007/s00521-012-1315-5 zbMATHGoogle Scholar
  10. 10.
    Yolcu U, Bas E, Egrioglu E, Aladag CH (2015) A new multilayer feed forward network model based on trimmed mean neuron model. Neural Netw World J 25:587–602CrossRefGoogle Scholar
  11. 11.
    Ghosh J, Shin Y (1992) Efficient higher-order neural networks for classification and function approximation. Int J Neural Syst 3:323–350CrossRefGoogle Scholar
  12. 12.
    Yadav RN, Kalra PK, John J (2007) Time series prediction with single multiplicative neuron model. Appl Soft Comput J 7:1157–1163. doi: 10.1016/j.asoc.2006.01.003 CrossRefGoogle Scholar
  13. 13.
    Chen W, Sun J, Gao S, Cheng J-J, Wang J, Todo Y (2017) Using a single dendritic neuron to forecast tourist arrivals to Japan. IEICE Trans Inf Syst E100D(1):190–202CrossRefGoogle Scholar
  14. 14.
    Zhou T, Ga S, Wang J, Chu C, Todo Y, Tang Z (2016) Financial time series prediction using a dendritic neuron model. Knowl-Based Syst 105:214–224CrossRefGoogle Scholar
  15. 15.
    Attia MA, Sallam EA, Fahmy MM. (2012) A proposed generalized mean single multiplicative neuron model. Proceedings—2012 I.E. 8th International Conference on Intelligent Computer Communication and Processing, ICCP 2012, art. no. 6356163, pp. 73–78Google Scholar
  16. 16.
    Bas E, Uslu VR, Egrioglu E (2016) Robust learning algorithm for multiplicative neuron model artificial neural networks. Expert Syst Appl 56:80–88. doi: 10.1016/j.eswa.2016.02.051 CrossRefGoogle Scholar
  17. 17.
    Barati M, Sharifian S (2015) A hybrid heuristic-based tuned support vector regression model for cloud load prediction. J Supercomput 71:4235–4259. doi: 10.1007/s11227-015-1520-y CrossRefGoogle Scholar
  18. 18.
    Berenguer TM, Berenguer JAM, García MEB et al (2015) Models of artificial neural networks applied to demand forecasting in nonconsolidated tourist destinations. Methodol Eur J Res Methods Behav Soc Sci 11:35–44. doi: 10.1027/1614-2241/a000088 Google Scholar
  19. 19.
    Haviluddin H (2015) Time series prediction using radial basis function neural network. Int J Electr Comput Eng 4:31–37CrossRefGoogle Scholar
  20. 20.
    Chow TWS, Cho SY (1997) Development of a recurrent Sigma-Pi neural network rainfall forecasting system in Hong Kong. Neural Comput Appl 5:66–75. doi: 10.1007/BF01501172 CrossRefGoogle Scholar
  21. 21.
    Cogollo MR, Velásquez JD (2014) Methodological advances in artificial neural networks for time series forecasting. IEEE Lat Am Trans 12:764–771. doi: 10.1109/TLA.2014.6868881 CrossRefGoogle Scholar
  22. 22.
    Thomas P, Suhner M-C (2015) A new multilayer perceptron pruning algorithm for classification and regression applications. Neural Process Lett 42:437–458. doi: 10.1007/s11063-014-9366-5 CrossRefGoogle Scholar
  23. 23.
    Beheshti Z, Firouzi M, Shamsuddin SM et al (2016) A new rainfall forecasting model using the CAPSO algorithm and an artificial neural network. Neural Comput Appl 27:2551–2565. doi: 10.1007/s00521-015-2024-7 CrossRefGoogle Scholar
  24. 24.
    Dey P, Sarkar A, Das AK (2016) Development of GEP and ANN model to predict the unsteady forced convection over a cylinder. Neural Comput Appl 27:2537–2549. doi: 10.1007/s00521-015-2023-8 CrossRefGoogle Scholar
  25. 25.
    Kiakojoori S, Khorasani K (2016) Dynamic neural networks for gas turbine engine degradation prediction, health monitoring and prognosis. Neural Comput Appl 27:2157–2192. doi: 10.1007/s00521-015-1990-0 CrossRefGoogle Scholar
  26. 26.
    Li H (2016) Research on prediction of traffic flow based on dynamic fuzzy neural networks. Neural Comput Appl 27:1969–1980. doi: 10.1007/s00521-015-1991-z CrossRefGoogle Scholar
  27. 27.
    Kennedy J, Eberhart R, Coello CAC, et al (1995) Particle swarm optimization. In: Neural Networks, 1995. Proceedings., IEEE Int. Conf. pp 1942–1948 vol.4Google Scholar
  28. 28.
    Shi Y, Eberhart RC (1999) Empirical study of particle swarm optimization. Evol Comput 1999 CEC 99 Proc 1999 Congr 3:1–1950 Vol. 3. doi:  10.1109/CEC.1999.785511
  29. 29.
    Yuchao M, Chuanwen J, Zhijian H, Chenming W (2006) The formulation of the optimal strategies for the electricity producers based on the particle swarm optimization algorithm. Power Syst IEEE Trans 21:1663–1671Google Scholar
  30. 30.
    Janacek G (2001) Practical time series. Oxford University Press Inc., Newyork, p 156zbMATHGoogle Scholar

Copyright information

© The Natural Computing Applications Forum 2017

Authors and Affiliations

  • Erol Egrioglu
    • 1
  • Ufuk Yolcu
    • 2
    Email author
  • Eren Bas
    • 1
  • Ali Zafer Dalar
    • 1
  1. 1.Department of Statistics, Forecast Research LaboratoryGiresun UniversityGiresunTurkey
  2. 2.Department of Econometrics, Forecast Research LaboratoryGiresun UniversityGiresunTurkey

Personalised recommendations