GenSoFNN-Yager: A novel brain-inspired generic self-organizing neuro-fuzzy system realizing Yager inference

摘要

Pattern recognition is increasingly becoming a key component of decision support systems (DSSs) in many application areas, especially when automatically extracting semantic rules from data is a chief concern. Accordingly, this paper presents a novel evolving neuro-fuzzy DSS, the generic self-organizing fuzzy neural network realizing Yager inference (GenSoFNN-Yager), that emulates the sequential learning paradigm of the hippocampus in the brain to synthesize from low-level numerical data to high-level declarative fuzzy rules. The proposed system exhibits simple and conceptually firm computational steps that correspond closely to a plausible human logical reasoning and decision-making. Experimental results on sample benchmark problems and realistic medical diagnosis applications show the potential of the proposed system as a competent DSS.