Fuzzy Logic Based Time Series Prediction Algorithm Using Nearest Neighborhood Clustering
Misbah Irum, Ejaz Muhammad · Journal of Engineering Research · 2020
This paper presents a fuzzy logic based prediction algorithm using nearest neigh-borhood clustering for the prediction of time series. The proposed algorithm hasbeen developed by using nearest neighborhood clustering scheme and optimalfuzzy system. The presented algorithm is tested by the prediction of Mackey-Glass time series, time series obtained from Logistic Map equation and an oilrefinery data series. To the best of our knowledge very few studies have at-tempted to predict Logistic Map. The results of the experimentation show thatthe presented prediction algorithm is able to predict linear, nonlinear and evenchaotic time series to a reasonable accuracy. It performes well even when themathematical model of time series does not exist.