Feature and architecture selection on deep feedforward network for roll motion time series prediction

Novri Suhermi*, Suhartono, Santi Puteri Rahayu, Fadilla Indrayuni Prastyasari, Baharuddin Ali, Muhammad Idrus Fachruddin

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

The neural architecture and the input features are very substantial in order to build an artificial neural network (ANN) model that is able to perform a good prediction. The architecture is determined by several hyperparameters including the number of hidden layers, the number of nodes in each hidden layer, the series length, and the activation function. In this study, we present a method to perform feature selection and architecture selection of ANN model for time series prediction. Specifically, we explore a deep learning or deep neural network (DNN) model, called deep feedforward network, an ANN model with multiple hidden layers. We use two approaches for selecting the inputs, namely PACF based inputs and ARIMA based inputs. Three activation functions used are logistic sigmoid, tanh, and ReLU. The real dataset used is time series data called roll motion of a Floating Production Unit (FPU). Root mean squared error (RMSE) is used as the model selection criteria. The results show that the ARIMA based 3 hidden layers DNN model with ReLU function outperforms with remarkable prediction accuracy among other models.

Original languageEnglish
Title of host publicationSoft Computing in Data Science - 4th International Conference, SCDS 2018, Proceedings
EditorsBee Wah Yap, Azlinah Hj Mohamed, Michael W. Berry
PublisherSpringer Verlag
Pages58-71
Number of pages14
ISBN (Print)9789811334405
DOIs
Publication statusPublished - 2019
Event4th International Conference on Soft Computing in Data Science, SCDS 2018 - Bangkok, Thailand
Duration: 15 Aug 201816 Aug 2018

Publication series

NameCommunications in Computer and Information Science
Volume937
ISSN (Print)1865-0929

Conference

Conference4th International Conference on Soft Computing in Data Science, SCDS 2018
Country/TerritoryThailand
CityBangkok
Period15/08/1816/08/18

Keywords

  • ARIMA
  • Deep feedforward network
  • PACF
  • Roll motion
  • Time series

Fingerprint

Dive into the research topics of 'Feature and architecture selection on deep feedforward network for roll motion time series prediction'. Together they form a unique fingerprint.

Cite this