journal article
A Neuro-Evolutive Interval Type-2 TSK Fuzzy System for Volatile Weather Forecasting
Fecha
2010Registro en:
10.1007/978-3-642-14922-1_19
Institución
Resumen
This paper presents an hybrid Neuro-Evolutive algorithm for a First-order Interval Type-2 TSK Fuzzy Logic System applied to a volatile weather forecasting case. All results are tested by statistical tests as Goldfeld-Quant, Ljung-Box, ARCH, Runs, Turning Points, Bayesian, Akaike and Hannan-Quin criteria. Some methodological aspects about a hybrid implementation among ANFIS, an Evolutive Optimizer and a First order Interval Type-2 TSK FLS are presented. The selected type-reduction algorithm is the IASCO algorithm proposed by Melgarejo in [1] since it presents better computing properties than other algorithms.