Remember Me
Or use your Academic/Social account:


Or use your Academic/Social account:


You have just completed your registration at OpenAire.

Before you can login to the site, you will need to activate your account. An e-mail will be sent to you with the proper instructions.


Please note that this site is currently undergoing Beta testing.
Any new content you create is not guaranteed to be present to the final version of the site upon release.

Thank you for your patience,
OpenAire Dev Team.

Close This Message


Verify Password:
Verify E-mail:
*All Fields Are Required.
Please Verify You Are Human:
fbtwitterlinkedinvimeoflicker grey 14rssslideshare1
Ruta, Dymitr; Gabrys, Bogdan (2007)
Publisher: IEEE Press
Languages: English
Types: Unknown
Subjects: aintel, csi
Rapidly evolving businesses generate massive\ud amounts of time-stamped data sequences and defy a demand\ud for massively multivariate time series analysis. For such data\ud the predictive engine shifts from the historical auto-regression\ud to modelling complex non-linear relationships between multidimensional\ud features and the time series outputs. In order to\ud exploit these time-disparate relationships for the improved time\ud series forecasting, the system requires a flexible methodology\ud of combining multiple prediction models applied to multiple\ud versions of the temporal data under significant noise component\ud and variable temporal depth of predictions. In reply\ud to this challenge a composite time series prediction model\ud is proposed which combines the strength of multiple neural\ud network (NN) regressors applied to the temporally varied\ud feature subsets and the postprocessing smoothing of outputs\ud developed to further reduce noise. The key strength of the model\ud is its excellent adaptability and generalisation ability achieved\ud through a highly diversified set of complementary NN models.\ud The model has been evaluated within NISIS Competition 2006\ud and NN3 Competition 2007 concerning prediction of univariate\ud and multivariate time-series. It showed the best predictive\ud performance among 12 competitive models in the NISIS 2006\ud and is under evaluation within NN3 2007 Competition.
  • The results below are discovered through our pilot algorithms. Let us know how we are doing!

    • [1] T. Dietterich and R. Michalski. Learning to predict sequences. In Machine Learning: An Artificial Intelligence Approach, vol. 2, Morgan Kaufmann, 1986
    • [2] R.S. Tsai, Analysis of financial time series, John Wiley & Sons, 2002.
    • [3] C.L. Giles, S. Lawrence and A.C. Tsoi, Noisy time series prediction using recurrent neural networks and grammatical inference, Machine Learning vol. 44(1/2), pp. 161-183, 2001.
    • [4] M. Casdagli, Nonlinear prediction of chaotic time series, Physica vol. 35, pp. 335-356, 1989.
    • [5] Z. Vojinovic, V. Kecman, R. Seidel, A data mining approach to financial time series modelling and forecasting, International Journal of Intelligent Systems in Accounting, Finance & Management vol. 10(4), pp. 225-239, 2001.
    • [6] F.E.H. Tay and L.J. Cao, Modified support vector machines in financial time series forecasting, Neurocomputing vol. 48(1), pp. 847-861, 2002.
    • [7] S. Mukherjee, E. Osuna and F. Girosi, Nonlinear prediction of chaotic time series using support vector machines, In Proc. of the 7th IEEE Workshop on Neural Networks for Signal Processing, Amelia Island, FL, IEEE Press, pp. 511-520, 1997.
    • [8] A.J.C Sharkey, Combining Artificial Neural Nets: Ensemble and Modular Multi-Net Systems. Springer-Verlag, London, UK, 1999.
    • [9] A.J.C Sharkey and N.E. Sharkey, Combining diverse neural nets, The Knowledge Engineering Review vol. 12(3), pp. 231-247, 1997.
    • [10] S. Hansen and P. Salamon, Neural network ensembles, IEEE Trans. on Patt. Analysis and Machine Intelligence 12(10), pp. 993-1001, 1990.
    • [11] C. Igel and M. Husken, Improving the Rprop Learning Algorithm, Proc. of the 2nd Int. ICSC Symposium on Neural Computation, pp. 115-121, 2000.
  • No related research data.
  • No similar publications.

Share - Bookmark

Cite this article