Fuzzy Cognitive Networks in Diverse Applications Using Hybrid Representative Structures

Abstract Fuzzy cognitive networks (FCNs) arose from traditional fuzzy maps (FCMs) to have the advantage of guaranteed convergence equilibrium points, thus being more suitable than conventional FCMs for a variety pattern recognition and system identification tasks. Moreover, recent developments led FCNs with functional weights (FCNs-FW), as significant enhancement in terms storage requirements, efficiency less human intervention requirements. In this paper we proceed further by introducing hybrid deep learning structures, interweaving FCNs-FW well established neural network (DNN) representative structures apply new schemes on time series prediction More specifically, after discussing general issues related construction using present three models, which combine FCN-FW convolutional (CNNs), echo state (ESNs) AutoEncoder (AE) schemes, respectively. The are tested diverse benchmark data sets prove that based perform equally or better state-of-the-art DNN-based paving way structures.