This paper validates a novel fuzzy classification methodology using the ELENA benchmark (ftp://ftp.dice.ucl.ac.be/ pub/neural-net/ELENA/databases). The main advantage of this methodology is the high accuracy with which it learns the topological structure of the features space. The fuzzy subsets built by the classifier approximate with a very small error the areas in the features space corresponding to different categories. Its accuracy also manifests through handling with fine precision the discrimination inside overlapping areas. These two properties of the fuzzy classifier have been confirmed by the testing results obtained on ELENA benchmark datasets.
Pattern recognition, Fuzzy logic, Fuzzy subsets, Clustering, Performance analysis.
All articles published by Studies in Informatics and Control (SIC) are published in accordance with the terms and conditions of the Creative Commons Attribution License (CC BY-NC-SA 4.0).