Title: A novel approach to classificatory problem using neuro-fuzzy architecture

Authors: Rahul Kala; Anupam Shukla; Ritu Tiwari

Addresses: Soft Computing and Expert System Laboratory, Indian Institute of Information Technology and Management, Gwalior, MP, India. ' Soft Computing and Expert System Laboratory, Indian Institute of Information Technology and Management, Gwalior, MP, India. ' Soft Computing and Expert System Laboratory, Indian Institute of Information Technology and Management, Gwalior, MP, India

Abstract: In this paper, we propose a new method for solving these problems inspired from the neuro-fuzzy logic approach for classificatory problems. We first cluster the training data based on class identification of inputs. A sort of fuzzy approach serves as a means to classify the unknown inputs. Rules are in the form of representative of every cluster and their matching class. The centre and power of the representative are the parameters that are optimised using a training algorithm and further by Genetic Algorithms. We tested the algorithm on the famous classificatory problem of picture learning.

Keywords: neuro-fuzzy networks; machine learning; clustering; fuzzy logic; artificial neural networks; ANNs; classification; pattern matching; genetic algorithms; picture learning.

DOI: 10.1504/IJSCC.2011.042432

International Journal of Systems, Control and Communications, 2011 Vol.3 No.3, pp.259 - 269

Published online: 11 Sep 2011 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article