• Jan 05, 2017 News![CFP] 2017 the annual meeting of IJFCC Editorial Board, ICCTD 2017, will be held in Paris, France during March 20-22, 2017.   [Click]
  • Mar 24, 2016 News! IJFCC Vol. 4, No. 4 has been indexed by EI (Inspec).   [Click]
  • Jun 28, 2017 News!Vol.6, No.3 has been published with online version.   [Click]
General Information
    • ISSN: 2010-3751
    • Frequency: Bimonthly (2012-2016); Quarterly (Since 2017)
    • DOI: 10.18178/IJFCC
    • Editor-in-Chief: Prof. Mohamed Othman
    • Executive Editor: Ms. Nancy Y. Liu
    • Abstracting/ Indexing: Google Scholar, Engineering & Technology Digital Library, and Crossref, DOAJ, Electronic Journals LibraryEI (INSPEC, IET).
    • E-mail:  ijfcc@ejournal.net 
Editor-in-chief
Prof. Mohamed Othman
Department of Communication Technology and Network Universiti Putra Malaysia, Malaysia
It is my honor to be the editor-in-chief of IJFCC. The journal publishes good papers in the field of future computer and communication. Hopefully, IJFCC will become a recognized journal among the readers in the filed of future computer and communication.
IJFCC 2017 Vol.6(3): 123-127 ISSN: 2010-3751
doi: 10.18178/ijfcc.2017.6.3.503

A Deep Learning of Time Series for Efficient Analysis

Tippaya Thinsungnoen, Kittisak Kerdprasop, and Nittaya Kerdprasop
Abstract—The main problem for analyzing time series data with machine learning techniques such as classification and clustering is that a high-dimensional nature of this kind of data can cause computational difficulty in finding optimal solution. Currently, advanced learning strategy such as deep learning has been used extensively and effectively to improve learning performance. In this research, we propose a method to optimize time series analysis by adding a pre-training and fine-tuning process of deep learning based on Deep Belief Networks and Restricted Boltzmann Machines. On evaluating performance of the proposed method, we use electroencephalographic, electrocardiogram, and synthetic time series data to analyze with classification task. The induced classification models are assessed with the four several metrics including cluster evaluation, purity, mean squared error, and processing time. We comparatively compare the three learning schemes: traditional neural networks, deep learning networks, and deep learning networks with added a pre-training and fine-tuning process. The results showed that all three schemes show the same performance on predicting time series data when assessed with mean squared error. For the processing time comparison, neural networks technique is slightly faster than others. But when assessed with cluster formation and purity metrics, we found that deep learning based on the concept of Deep Belief Networks and Restricted Boltzmann Machines that adds a pre-training and fine-tuning process outperforms other learning techniques.

Index Terms—Deep belief networks, deep learning, restricted Boltzmann machines, time series analysis.

The authors are with the School of Computer Engineering, Suranaree University of Technology (SUT), 111 University Avenue, Muang, Nakhon Ratchasima 30000, Thailand (corresponding author: T. Thinsungnoen; Tel.: +66819671907; e-mail: tippayasot@hotmail.com, kerdpras@sut.ac.th, nittaya@sut.ac.th).

[PDF]

Cite: Tippaya Thinsungnoen, Kittisak Kerdprasop, and Nittaya Kerdprasop, "A Deep Learning of Time Series for Efficient Analysis," International Journal of Future Computer and Communication vol. 6, no. 3, pp. 123-127, 2017.

Copyright © 2008-2016. International Journal of Future Computer and Communication. All rights reserved.
E-mail: ijfcc@ejournal.net