Automatic Generation of a Type-2 Fuzzy System for Time Series Forecast based on Genetic Programming

摘要

This work describes the development of a type 2 Fuzzy Inference System by using Genetic Programming for applications in time series forecasting. The resulting model, called GPFIS-Forecast+, is based on the GPFIS-Forecast created previously, which made use of Multigene Genetic Programming an provided good results. Results show that the system with type 2 fuzzy sets improves the performance, especially for noisy data.