In this work we discuss an approach for deriving fuzzy classifier models directly from data. The aim is to find a fuzzy classifier with high performance and low complexity, so as it can be easily interpretable. To do this, we first extract a raw model of a fuzzy classifier from training examples through a simple procedure, and then refine the model by using a neuro-fuzzy approach. Accuracy is improved by means of a learning technique, and interpretability is obtained by reducing the rule base. The proposed approach is verified on the Iris classification problem.
Scheda prodotto non validato
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo
Titolo: | Fuzzy classifiers acquired from data |
Autori: | |
Data di pubblicazione: | 2000 |
Handle: | http://hdl.handle.net/11586/6507 |
ISBN: | 90-5199-476-1 |
Appare nelle tipologie: | 2.1 Contributo in volume (Capitolo o Saggio) |