Fuzzy ART Combining Overlapped Categories Using Variable Vigilance Parameters
Haruka Isawa, Haruna Matsushita, Yoshifumi Nishio · 2009
The Fuzzy Adaptive Resonance Theory (Fuzzy ART) is an unsupervised neural network and allows both binary and continuous input patterns. In this study, we propose a Fuzzy ART Combining Overlapped Categories Using Variable Vigilance Parameters. The vigilance parameters of the proposed method are arranged for every category, and they are varied according to the size of respective categories with learning. We confirm that the proposed Fuzzy ART can classify input data more flexible than the conventional Fuzzy ART. 1.