Detail publikace

Comparison of Neural Network and Grammatical Evolution for Time Series Prediction

LÝSEK, J. ŠŤASTNÝ, J. MOTYČKA, A.

Anglický název

Comparison of Neural Network and Grammatical Evolution for Time Series Prediction

Typ

Stať ve sborníku v databázi WoS či Scopus

Jazyk

en

Originální abstrakt

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.

Klíčová slova anglicky

neural network, grammatical evolution, learning, prediction

Vydáno

2013-06-26

Místo

Brno

ISBN

978-80-214-4755-4

Kniha

Mendel 2013

Strany od–do

215–220

Počet stran

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"
}