Title :
A Novel Granular Neural Network Architecture
Author :
Dick, S. ; Tappenden, A. ; Badke, Curtis ; Olarewaju, O.
Author_Institution :
Univ. of Alberta, Edmonton
Abstract :
We introduce a novel granular neural network (GNN) architecture based on the multi-layer perceptron architecture. The GNN uses linguistic terms as connection weights, and uses the operations of linguistic arithmetic to update those connection weights. The GNN has been implemented in a Java-based simulation environment, with support for both regression and classification learning tasks. We present the results of a preliminary experimental comparison between the GNN and the c4.5 decision tree algorithm on two benchmark datasets. Our results show that the GNN was slightly more accurate than c4.5 on both datasets.
Keywords :
Java; decision trees; learning (artificial intelligence); multilayer perceptrons; neural net architecture; regression analysis; Java-based simulation environment; c4.5 decision tree algorithm; classification learning task; connection weight; granular neural network architecture; linguistic arithmetic; multilayer perceptron architecture; regression task; Arithmetic; Computer architecture; Computer networks; Fuzzy sets; Fuzzy systems; Java; Machine learning; Multi-layer neural network; Neural networks; Set theory; Granular computing; Granular neural networks; Machine learning; Neuro-fuzzy systems;
Conference_Titel :
Fuzzy Information Processing Society, 2007. NAFIPS '07. Annual Meeting of the North American
Conference_Location :
San Diego, CA
Print_ISBN :
1-4244-1213-7
Electronic_ISBN :
1-4244-1214-5
DOI :
10.1109/NAFIPS.2007.383808