Title :
Neural network expert system
Author :
Galushkin, A.I. ; Savushkin, S.A.
Author_Institution :
Acad. of Sci., Moscow, Russia
Abstract :
There are two approaches to applying neural mathematics in the expert system (ES) field: realization of neural ES technology and neural realization of modules of traditional ES technology. The first approach requires a great deal of systematization of the problem area, which may maintain a finite (perhaps very large) number of objects and relations. The neural ES implements traditional ES functions. It is possible to create a self-learning ES based on Samual´s idea. Attention is given to a system processing incomplete information, a neural network system of knowledge processing and acquisition, a Boolean-logic-based ES, and the synthesis of a network system with varying structures
Keywords :
Boolean functions; expert systems; knowledge acquisition; neural nets; unsupervised learning; Boolean logic; expert system; incomplete information; knowledge acquisition; knowledge processing; neural mathematics; neural network system; self-learning; Artificial neural networks; Chemistry; Data mining; Engines; Expert systems; Knowledge engineering; Medical diagnosis; Medical expert systems; Neural networks;
Conference_Titel :
Neuroinformatics and Neurocomputers, 1992., RNNS/IEEE Symposium on
Conference_Location :
Rostov-on-Don
Print_ISBN :
0-7803-0809-3
DOI :
10.1109/RNNS.1992.268517
