A new neural network architecture that consists of neo-fuzzy-neurons and a learning algorithm that is characterized with high speed and filtering properties are proposed. The proposed architecture can be used for real-time short-term forecasting.
adaptive prognostication, neo-fuzzy-neuron, multidimensional non-stationary temporal rows, teaching algorithm
"Adaptyvnыi neo-fazzy-predyktor mnohomernыkh nestatsyonarnыkh vremennыkh riadov" ,
Information Processing Systems,