A comparative study of general fuzzy min-max neural networks for pattern classification problems

• Published in: Neurocomputing (Amsterdam) Vol. 386; pp. 110 - 125
• Main Authors: Khuat, Thanh Tung, Gabrys, Bogdan
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 21.04.2020
• Subjects: Comparative study; Fuzzy min-max neural network; General fuzzy min-max; Hyperbox fuzzy sets; Pattern classification; Hyperbox fuzzy sets; General fuzzy min-max; Fuzzy min-max neural network; Pattern classification; Comparative study
• ISSN: 0925-2312; 1872-8286
• DOI: 10.1016/j.neucom.2019.12.090

•Identify advantages and drawbacks of general fuzzy min-max neural network (GFMM).•Empirically evaluate the GFMM in comparison to other machine learning algorithms.•Present the roles of robust evaluation techniques in analyzing empirical results.•Propose solutions for classification problems with small-sized training sets.•Identify existing issues and inform potential research directions for GFMM. General fuzzy min-max (GFMM) neural network is a generalization of fuzzy neural networks formed by hyperbox fuzzy sets for classification and clustering problems. Two principle algorithms are deployed to train this type of neural network, i.e., incremental learning and agglomerative learning. This paper presents a comprehensive empirical study of performance influencing factors, advantages, and drawbacks of the general fuzzy min-max neural network on pattern classification problems. The subjects of this study include (1) the impact of maximum hyperbox size, (2) the influence of the similarity threshold and measures on the agglomerative learning algorithm, (3) the effect of data presentation order, (4) comparative performance evaluation of the GFMM with other types of fuzzy min-max neural networks and prevalent machine learning algorithms. The experimental results on benchmark datasets widely used in machine learning showed overall strong and weak points of the GFMM classifier. These outcomes also informed potential research directions for this class of machine learning algorithms in the future.