A type-2 fuzzy neural network ensemble to predict chaotic time series

Abstract

This paper presents a type-2 adaptive fuzzy neural network ensemble to predict chaotic time series in combination with the well known M8 algorithm. The chaotic time series is depicted by the register of seismic events and their seismic coordinates in a catalog. ANFIS model are used as components of the Ensemble to train and evaluate seven chaotic time series that are used by the M8 algorithm to make a prediction.