Pre-processing inputs for optimally-configured time-delay neural networks

TAŞKAYA TEMİZEL T., Casey M., Ahmad K.

ELECTRONICS LETTERS, vol.41, no.4, pp.198-200, 2005 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 41 Issue: 4
  • Publication Date: 2005
  • Doi Number: 10.1049/el:20058016
  • Journal Name: ELECTRONICS LETTERS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.198-200
  • Middle East Technical University Affiliated: No

Abstract

A procedure for pre-processing non-stationary time series is proposed for modelling with a time-delay neural network (TDNN). The procedure stabilises the mean of the series and uses a fast Fourier transform to determine the TDNN input size. Results of applying this procedure on five well-known data sets are compared with existing hybrid neural network techniques, demonstrating improved prediction performance.