Title :
Learning in certainty-factor-based neural networks
Author_Institution :
Dept. of Comput. & Inf. Sci., Florida Univ., Gainesville, FL
Abstract :
The certainty-factor-based neural network refers to a multilayer neural network where the network activation function is based on the certainty factor (CF) model of MYCIN-like systems. It is shown that the neural network using the CF-based activation function requires relatively small sample sizes for correct generalization and hence also facilitates learning rules. These findings are confirmed by empirical studies. Experiments suggest that the CFNet is capable of discovering the underlying domain rules
Keywords :
case-based reasoning; generalisation (artificial intelligence); learning (artificial intelligence); multilayer perceptrons; transfer functions; MYCIN-like systems; certainty-factor-based neural networks; domain rules; generalization; learning rules; multilayer neural network; network activation function; Algorithm design and analysis; Artificial intelligence; Computer networks; Intelligent networks; Learning systems; Multi-layer neural network; Neural networks; Neurons; Upper bound;
Conference_Titel :
Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on
Conference_Location :
Anchorage, AK
Print_ISBN :
0-7803-4859-1
DOI :
10.1109/IJCNN.1998.682234
