METHODS OF BUILDING OF TRAINING SAMPLE IN TASKS OF TIME SERIES MODELING BY NEURAL NETWORKS

Authors

  • Roman Naumovych Kvietnyi Vinnitsa National Technical University, Vinnitsa, Khmelnytske Highway 95, (0432) 598 243
  • Olha Olehivna Druzhynina Vinnitsa National Technical University, Vinnitsa, Khmelnytske Highway 95, (0432) 598 243

Abstract

Existing approaches to the training set generating issue for time series modeling using radial-basis neural networks tasks were analyzed. Methods of training set features preprocessing are considered. Experimental results confirm the reasonability of additional input components using and input data preprocessing.

Author Biographies

Roman Naumovych Kvietnyi, Vinnitsa National Technical University, Vinnitsa, Khmelnytske Highway 95, (0432) 598 243

d.t.s., proff.,, Head of the Department of AIIT

Olha Olehivna Druzhynina, Vinnitsa National Technical University, Vinnitsa, Khmelnytske Highway 95, (0432) 598 243

graduate student of the Department of AIIT

Downloads

Abstract views: 255

How to Cite

[1]
R. N. Kvietnyi and O. O. Druzhynina, “METHODS OF BUILDING OF TRAINING SAMPLE IN TASKS OF TIME SERIES MODELING BY NEURAL NETWORKS”, ІТКІ, vol. 23, no. 1, Oct. 2012.

Issue

Vol. 23 No. 1 (2012): Information Technology and Computer Engineering

Section

Mathematical modeling and computational methods

Metrics

Downloads

Download data is not yet available.

Most read articles by the same author(s)

1 2 > >>