FSR-ARTMAP: a modified fuzzy ARTMAP with fuzzy signals and rules

摘要

Introduces a neural architecture termed the "Fuzzy Signals and Rules ARTMAP" (FSR-ARTMAP). It is a modified fuzzy ARTMAP for incremental supervised learning and multidimensional mapping using fuzzy signals. This neural architecture is able to process and store linguistic information in the form of fuzzy logic rules and membership functions for fuzzy inference. The FSR-ARTMAP architecture is composed of two similar "Fuzzy Signals and Rules ART" (FSR-ART) modules and a mapping field. FSR-ART is a modified version of fuzzy ART. The mapping field between the input and the output FSR-ART modules can be one-to-one, many-to-one or one-to-many.