A new approach for probability calculation of fuzzy events in Bayesian Networks


İÇEN D. , ERSEL D.

INTERNATIONAL JOURNAL OF APPROXIMATE REASONING, cilt.108, ss.76-88, 2019 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 108
  • Basım Tarihi: 2019
  • Doi Numarası: 10.1016/j.ijar.2019.03.004
  • Dergi Adı: INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
  • Sayfa Sayıları: ss.76-88

Özet

In this study, we propose an approach that incorporates the advantages of fuzzy events and fuzzy probabilities, as they have not been used together before in the literature for Bayesian Networks. We focus on the fuzzy events in Bayesian Networks and calculate their fuzzy marginal and fuzzy conditional probabilities through Buckley's fuzzy probability method. We intend to show that this approach can be used for achieving more interpretable results for real life problems and represent the uncertainty better. Moreover, detailed information in the data can be used more effectively. (C) 2019 Elsevier Inc. All rights reserved.