TRANSFORMATION OF AN INFERENCE TREE TO A LOGICALLY EQUIVALENT NEURAL NETWORK
DOI:
https://doi.org/10.17770/etr1997vol1.1853Abstract
III-formalized situations can easily be represented by means of conditional rules. Heuristics that are represented through rules well describe the relationships among the factors of the process or event to be examined. , .
In such a situation two tasks should usually be solved. The first deals with constructing the logical structure of a set of rules which properly represent the described process. The second task is the adjustment of this structure on the level of quantitative evaluations which characterize the described relationships. A logically full, consistent and non-exhaustive set of rules is adjusted through learning procedures taken from the apparatus of artificial neural networks (Zurada 1992).
When implementing this approch, the following tasks arise that are considered in the present:
(i) construction of the adequate neuronlike structure,
(ii) choice and implementation of a learning algorithm for training the above structure.
Other investigations are also known that have used a neural network as an element of an expert system. Gallant (1986) proposes the methodology of artificial neural network application for the task of knowledge acquisition task.
Downloads
References
Gallant S. I. Connectionist expert systems. Communications of the ACM. Vol.31. - N2. - 1988. - 153. - 169. p.
Harmon P. and King D. Expert Systems: Artificial Intelligence in Business. J.Wiley (Ed.). - New York, 1985.
Li Miri Fu. Knowledge-based connectionism for revising domain theories. IEEE Transactions on Systems, Man and Cybernetics. Vol. 23. - N 1.- 1993. - 173. - 182, p.
Rumeihart D. E., Hinton G. E. and Williams R. J. Learning internal representation by error propagation. Parallel Distributed Processing.
Rumelhart and McVelloud (Eds). - Vol. 1,2.- Massachusetts: MIT Press, 1986.-318.-362.p,
Tan C. L., Quah T. S. and Teh H. H. An artificial neural network that models human decision making. IEEE Computer Society, 1996. - 64. - 70. p.
Zurada J. M. Introduction to Artificial Neural Systems. - St.Paul: West Publishing Company, 1992. - 683 p.
Nelson M. and Illing Worth W. T. A Practical Guide to Neural Nets. - Addison: Wesley Publishing Company, 1991.
