A Comparison between Deep Belief Network and LSTM in Chaotic Time Series Forecasting
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F21%3A10249628" target="_blank" >RIV/61989100:27240/21:10249628 - isvavai.cz</a>
Result on the web
<a href="https://dl.acm.org/doi/pdf/10.1145/3490725.3490749" target="_blank" >https://dl.acm.org/doi/pdf/10.1145/3490725.3490749</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1145/3490725.3490749" target="_blank" >10.1145/3490725.3490749</a>
Alternative languages
Result language
angličtina
Original language name
A Comparison between Deep Belief Network and LSTM in Chaotic Time Series Forecasting
Original language description
The application of deep neural networks in forecasting time series data is increasingly popular, aiming to improve prediction accuracy in this problem. However, as for chaotic time series, a special kind of time series data generated from the deterministic dynamics of a nonlinear system, there are very few research works applying deep neural networks to forecast this kind of time series data. So far, Deep Belief Network (DBN) and Long Short Term Memory (LSTM) are two kinds of deep neural networks used to extract meaningful features from the chaotic time series before forecasting. This article aims to compare the prediction performance of the LSTM model with that of the DBN model on chaotic time series data. Experimental results on six synthetic and real-world datasets in this study show that LSTM brings out better prediction accuracy than DBN in terms of three evaluation criteria. (C) 2021 ACM.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
—
Continuities
S - Specificky vyzkum na vysokych skolach
Others
Publication year
2021
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
ACM International Conference Proceeding Series 2021
ISBN
978-1-4503-8424-7
ISSN
—
e-ISSN
—
Number of pages
7
Pages from-to
157-163
Publisher name
Association for Computing Machinery
Place of publication
New York
Event location
Chang-čou
Event date
Sep 17, 2021
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—