Neural Processing Letters

, Volume 50, Issue 2, pp 1789–1829 | Cite as

Several Novel Dynamic Ensemble Selection Algorithms for Time Series Prediction

  • Changsheng Yao
  • Qun DaiEmail author
  • Gang Song


The goal to improve prediction accuracy and robustness of predictive models is quite important for time series prediction (TSP). Multi-model predictions ensemble exhibits favorable capability to enhance forecasting precision. Nevertheless, a static ensemble system does not always function well for all the circumstances. This work proposes six novel dynamic ensemble selection (DES) algorithms for TSP, including one DES algorithm based on Predictor Accuracy over Local Region (DES-PALR), two DES algorithms based on the Consensus of Predictors (DES-CP) and three Dynamic Validation Set determination algorithms. The first dynamic validation set determination algorithm is designed based on the similarity between the Predictive value of the test sample and the Objective values of the training samples. The second one is constructed based on the similarity between the Newly constituted sample for the test sample and All the training samples. Finally, the third one is developed based on the similarity between the Output profile of the test sample and the Output profile of each training sample. These proposed algorithms successfully realize dynamic ensemble selection for TSP. Experimental results on twelve benchmark time series datasets have demonstrated that the proposed DES algorithms greatly improve predictive performance when compared against current state-of-the-art prediction algorithms and the static ensemble selection techniques.


Dynamic ensemble selection (DES) DES algorithm based on Predictor Accuracy over Local Region (DES-PALR) Dynamic Ensemble Selection algorithm based on the Consensus of Predictors (DES-CP) Dynamic validation set determination algorithm Time series prediction (TSP) 



This work is supported by the National Natural Science Foundation of China under the Grant No. 61473150.


  1. 1.
    Hamilton JD (1994) Time series analysis. Princeton University Press, PrincetonzbMATHGoogle Scholar
  2. 2.
    Brockwell PJ, Davis RA (2009) Introduction to time series and forecasting. Springer, BerlinzbMATHGoogle Scholar
  3. 3.
    Pulido ME, Melin P (2012) Optimization of type-2 fuzzy integration in ensemble neural networks for predicting the Dow Jones time series. In: Fuzzy information processing society, pp 1–6Google Scholar
  4. 4.
    Palivonaite R, Ragulskis M (2016) Short-term time series algebraic forecasting with internal smoothing. Neurocomputing 171:854–865CrossRefGoogle Scholar
  5. 5.
    Ma Z, Dai Q (2016) Selected an stacking ELMs for time series prediction. Neural Process Lett 44:1–26CrossRefGoogle Scholar
  6. 6.
    Balkin SD, Ord JK (2000) Automatic neural network modeling for univariate time series. Int J Forecast 16:509–515CrossRefGoogle Scholar
  7. 7.
    Giordano F, La Rocca M, Perna C (2007) Forecasting nonlinear time series with neural network sieve bootstrap. Comput Stat Data Anal 51:3871–3884MathSciNetCrossRefGoogle Scholar
  8. 8.
    Jain A, Kumar AM (2007) Hybrid neural network models for hydrologic time series forecasting. Appl Soft Comput 7:585–592CrossRefGoogle Scholar
  9. 9.
    Lapedes AS, Farber RF (1987) Nonlinear signal processing using neural networks: prediction and system modeling. In: 1. IEEE international conference on neural networksGoogle Scholar
  10. 10.
    Chakraborty K, Mehrotra K, Mohan CK, Ranka S (1992) Original contribution: forecasting the behavior of multivariate time series using neural networks. Neural Netw 5:961–970CrossRefGoogle Scholar
  11. 11.
    Chatfield C, Weigend AS (1994) Time series prediction: forecasting the future and understanding the past: Neil A. Gershenfeld and Andreas S. Weigend, 1994, ‘The future of time series’, in: A.S. Weigend and N.A. Gershenfeld, eds., (Addison-Wesley, Reading, MA), 1–70. Int J Forecast 10:161–163CrossRefGoogle Scholar
  12. 12.
    Adhikari R (2015) A neural network based linear ensemble framework for time series forecasting. Neurocomputing 157:231–242CrossRefGoogle Scholar
  13. 13.
    Pelikan E, Groot CD, Wurtz D (1992) Power consumption in West-Bohemia: improved forecasts with decorrelating connectionist networks. Neural Netw World 2:701–712Google Scholar
  14. 14.
    Britto AS, Sabourin R, Oliveira LES (2014) Dynamic selection of classifiers—a comprehensive review. Pattern Recognit 47:3665–3680CrossRefGoogle Scholar
  15. 15.
    Kittler J, Hatef M, Duin RPW, Matas J (1998) On combining classifiers. IEEE Trans Pattern Anal Mach Intell 20:226–239CrossRefGoogle Scholar
  16. 16.
    Adhikari R, Verma G, Khandelwal I (2014) A model ranking based selective ensemble approach for time series forecasting. In: International conference on intelligent computing, communication and convergence, pp 14–21Google Scholar
  17. 17.
    Cruz RMO, Sabourin R, Cavalcanti GDC, Ren TI (2015) META-DES: a dynamic ensemble selection framework using meta-learning. Pattern Recognit 48:1925–1935CrossRefGoogle Scholar
  18. 18.
    Gheyas IA, Smith LS (2011) A novel neural network ensemble architecture for time series forecasting. Neurocomputing 74:3855–3864CrossRefGoogle Scholar
  19. 19.
    Kourentzes N, Barrow DK, Crone SF (2014) Neural network ensemble operators for time series forecasting. Expert Syst Appl Int J 41:4235–4244CrossRefGoogle Scholar
  20. 20.
    Donate JP, Cortez P, Sánchez GG, Miguel ASD (2013) Time series forecasting using a weighted cross-validation evolutionary artificial neural network ensemble. Neurocomputing 109:27–32CrossRefGoogle Scholar
  21. 21.
    Krikunov AV, Kovalchuk SV (2015) Dynamic selection of ensemble members in multi-model hydrometeorological ensemble forecasting. Procedia Comput Sci 66:220–227CrossRefGoogle Scholar
  22. 22.
    Adhikari R, Verma G (2016) Time series forecasting through a dynamic weighted ensemble approach. Springer, New DelhiCrossRefGoogle Scholar
  23. 23.
    Kolter JZ, Maloof MA (2007) Dynamic weighted majority: an ensemble method for drifting concepts. J Mach Learn Res 8:2755–2790zbMATHGoogle Scholar
  24. 24.
    Woods K, Kegelmeyer WP, Bowyer K (1997) Combination of multiple classifiers using local accuracy estimates. IEEE Trans Pattern Anal Mach Intell 19:405–410CrossRefGoogle Scholar
  25. 25.
    Smits PC (2002) Multiple classifier systems for supervised remote sensing image classification based on dynamic classifier selection. IEEE Trans Geosci Remote Sens 40(4):801–813CrossRefGoogle Scholar
  26. 26.
    Kuncheva LI (2000) Clustering-and-selection model for classifier combination. In: International conference on knowledge-based intelligent engineering systems and allied technologies. Proceedings, vol 1, pp 185–188Google Scholar
  27. 27.
    Kuncheva LI, Bezdek JC, Duin RPW (2001) Decision templates for multiple classifier fusion: an experimental comparison. Pattern Recognit 34:299–314CrossRefGoogle Scholar
  28. 28.
    Zhou ZH, Wu JX, Jiang Y, Chen SF (2001) Genetic algorithm based selective neural network ensemble. In: International joint conference on artificial intelligence, pp 797–802Google Scholar
  29. 29.
    Santos EMD, Sabourin R, Maupin P (2008) A dynamic overproduce-and-choose strategy for the selection of classifier ensembles. Pattern Recognit 41:2993–3009CrossRefGoogle Scholar
  30. 30.
    Huang GB, Zhu QY, Siew CK (2006) Extreme learning machine: theory and applications. Neurocomputing 70:489–501CrossRefGoogle Scholar
  31. 31.
    Zhao G, Shen Z, Miao C, Gay R (2008) Enhanced Extreme learning machine with stacked generalization. In: International joint conference on neural networks, pp 1191–1198Google Scholar
  32. 32.
    Huang GB (2003) Learning capability and storage capacity of two-hidden-layer feedforward networks. IEEE Trans Neural Netw 14:274–281CrossRefGoogle Scholar
  33. 33.
    Huang GB, Zhu QY, Siew CK (2004) Extreme learning machine: a new learning scheme of feedforward neural networks. In: IEEE international joint conference on neural networks, vol 2, pp 985–990Google Scholar
  34. 34.
    Zhou Z-H, Wu J, Tang W (2002) Ensembling neural networks: many could be better than all. Artif Intell 137:239–263MathSciNetCrossRefGoogle Scholar
  35. 35.
    Ko AHR, Sabourin R, Britto AS (2008) From dynamic classifier selection to dynamic ensemble selection. Pattern Recognit 41:1718–1731CrossRefGoogle Scholar
  36. 36.
    Kuncheva LI (2002) Switching between selection and fusion in combining classifiers: an experiment. IEEE Trans Syst Man Cybern B Cybern 32:146–156CrossRefGoogle Scholar
  37. 37.
    Hartigan JA, Wong MA (1979) Algorithm AS 136: a k-means clustering algorithm. Appl Stat 28:100–108CrossRefGoogle Scholar
  38. 38.
    Wang S, Qi L, Yu P, Peng X (2011) CLS-SVM: a local modeling method for time series forecasting. Chin J Sci Instrum 32:1824–1829Google Scholar
  39. 39.
    Paterlini S, Minerva T (2003) Evolutionary approaches for cluster analysis. Soft Computing Applications. Physica-Verlag, HDGoogle Scholar
  40. 40.
    Bezdek JC, Pal NR (1998) Some new indexes of cluster validity. IEEE Trans Syst Man Cybern B Cybern 28:301–315CrossRefGoogle Scholar
  41. 41.
    Dos Santos EM, Sabourin R, Maupin P (2006) Single and multi-objective genetic algorithms for the selection of ensemble of classifiers. In: International joint conference on neural networks, IJCNN, pp 3070–3077Google Scholar
  42. 42.
    Partalas I, Tsoumakas G, Vlahavas I (2008) Focused ensemble selection: a diversity-based method for greedy ensemble selection. In: European Conference on Artificial Intelligence (ECAI). IOS PressGoogle Scholar
  43. 43.
    Goldberg DE (1989) Genetic algorithms in search, optimization, and machine learning. Addison-Wesley Pub. Co, BostonzbMATHGoogle Scholar
  44. 44.
    Hyndman R (ed) Time Series Data Library.
  45. 45.
  46. 46.

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.College of Computer Science and TechnologyNanjing University of Aeronautics and AstronauticsNanjingChina

Personalised recommendations