Using Artificial Neural Networks in Fuzzy Time Series Analysis


Aladag C. H.

RECENT DEVELOPMENTS AND NEW DIRECTIONS IN SOFT COMPUTING, vol.317, pp.443-451, 2014 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 317
  • Publication Date: 2014
  • Doi Number: 10.1007/978-3-319-06323-2_28
  • Title of Journal : RECENT DEVELOPMENTS AND NEW DIRECTIONS IN SOFT COMPUTING
  • Page Numbers: pp.443-451

Abstract

In recent years, fuzzy time series have been drawn great attention due to their potential for use in time series forecasting. In many studies available in the literature, fuzzy time series have been successfully used to forecast time series contain some uncertainty. Studies on this method still continue to reach better forecasting results. Determination of fuzzy relations between observations is an important phase of fuzzy time series approaches which directly affect the forecasting performance. In order to establish fuzzy relations, different techniques have been utilized in the literature. One of these techniques is artificial neural networks method. In this study, it is shown that how different artificial neural networks models can be used to determine fuzzy relations with real time series applications.