Vague and incomplete data represented as linguistic values massively exists in diverse real-word applications. The task of forecasting fuzzy time series under uncertain circumstances is thus of great important but difficult. The inherent uncertainty involving time evolution usually makes the transition of states in a system probabilistic. In this paper, we proposed a new forecasting model based on Hidden Markov Model for fuzzy time series to realize the probabilistic state transition. We conduct experiments of forecasting a real-world temperature application to validate the better accuracy of the proposed model achieved over traditional fuzzy time series models.
|頁（從 - 到）||1919-1924|
|期刊||WSEAS Transactions on Systems|
|出版狀態||Published - 2006 八月|
All Science Journal Classification (ASJC) codes