Publication detail

Comparison of Neural Network and Grammatical Evolution for Time Series Prediction

LÝSEK, J. ŠŤASTNÝ, J. MOTYČKA, A.

Original Title

Comparison of Neural Network and Grammatical Evolution for Time Series Prediction

Type

conference paper

Language

English

Original Abstract

In this contribution we present a comparison of a neural network with an agent created by grammatical evolution for time series prediction problem. We use the back-propagation algorithm to train the neural network for predicting future value of a time series. To evolve prediction agents we use grammatical evolution with backwards processing and equalization operator. Finally the results of both methods are compared by their ability to predict time series values, by time needed to create prediction model and setup difficulty.

Keywords

neural network, grammatical evolution, learning, prediction

Key words in English

neural network, grammatical evolution, learning, prediction

Authors

LÝSEK, J.; ŠŤASTNÝ, J.; MOTYČKA, A.

RIV year

2013

Released

26. 6. 2013

Location

Brno

ISBN

978-80-214-4755-4

Book

Mendel 2013

Pages from

215

Pages to

220

Pages count

6

BibTex

@inproceedings{BUT100945,
  author="Jiří {Lýsek} and Jiří {Šťastný} and Arnošt {Motyčka}",
  title="Comparison of Neural Network and Grammatical Evolution for Time Series Prediction",
  booktitle="Mendel 2013",
  year="2013",
  pages="215--220",
  address="Brno",
  isbn="978-80-214-4755-4"
}