Abstract
A prototypical connectionist expert system is presented. DESKNET is built for ultimate use in the instruction of medical students to diagnose papulosquamous skin diseases. DESKNET is an ES composed of a three layered back propagation network consisting of an input layer, a hidden layer, and an output layer. The units on a layer have excitatory or inhibitory connections to units on the adjacent layers. Input for the system is a list of symptoms organized by 18 parameters. The back propagation algorithm encodes the implicit 'knowledge' such that it remains in an implicit form; thus, more knowledge can be captured. Therefore, the back propagation model appears to be a useful tool for representing concepts of a domain and encoding them in a network. Limitations of the algorithm are identified.
Original language | English (US) |
---|---|
Pages (from-to) | 477 |
Number of pages | 1 |
Journal | Neural Networks |
Volume | 1 |
Issue number | 1 SUPPL |
DOIs | |
State | Published - 1988 |
Event | International Neural Network Society 1988 First Annual Meeting - Boston, MA, USA Duration: Sep 6 1988 → Sep 10 1988 |
ASJC Scopus subject areas
- Cognitive Neuroscience
- Artificial Intelligence