Intuitionistic fuzzy time series functions approach for time series forecasting

Creative Commons License

Bas E., Yolcu U., Egrioglu E.

GRANULAR COMPUTING, cilt.6, sa.3, ss.619-629, 2021 (ESCI) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 6 Sayı: 3
  • Basım Tarihi: 2021
  • Doi Numarası: 10.1007/s41066-020-00220-8
  • Dergi Adı: GRANULAR COMPUTING
  • Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus
  • Sayfa Sayıları: ss.619-629
  • Anahtar Kelimeler: Intuitionistic fuzzy sets, Fuzzy inference, Forecasting, Fuzzy functions approach, INFERENCE SYSTEM, MODEL, RULES, ENROLLMENTS, ALGORITHM, ANFIS
  • Marmara Üniversitesi Adresli: Evet

Özet

Fuzzy inference systems have been commonly used for time series forecasting in the literature. Adaptive network fuzzy inference system, fuzzy time series approaches and fuzzy regression functions approaches are popular among fuzzy inference systems. In recent years, intuitionistic fuzzy sets have been preferred in the fuzzy modeling and new fuzzy inference systems have been proposed based on intuitionistic fuzzy sets. In this paper, a new intuitionistic fuzzy regression functions approach is proposed based on intuitionistic fuzzy sets for forecasting purpose. This new inference system is called an intuitionistic fuzzy time series functions approach. The contribution of the paper is proposing a new intuitionistic fuzzy inference system. To evaluate the performance of intuitionistic fuzzy time series functions, twenty-three real-world time series data sets are analyzed. The results obtained from the intuitionistic fuzzy time series functions approach are compared with some other methods according to a root mean square error and mean absolute percentage error criteria. The proposed method has superior forecasting performance among all methods.