Artificial neural networks (ANN) have been demonstrated to be increasi
ngly more useful for complex problems difficult to solve with conventi
onal methods. With their learning abilities, they avoid having to deve
lop a mathematical model or acquiring the appropriate knowledge to sol
ve a task. The difficulty now lies in the ANN design process. A lot of
choices must be made to design an ANN, and there are no available des
ign rules to make these choices directly for a particular problem. The
refore, the design of an ANN demands a certain number of iterations, m
ainly guided by the expertise and the intuition of the developer. To a
utomate the ANN design process, we have developed Neurex, composed of
an expert system and an ANN simulator. Neurex autonomously guides the
iterative ANN design process. Its structure tries to reproduce the des
ign steps done by a human expert in conceiving an ANN. As a whole, the
Neurex structure serves as a framework to implement this expertise fo
r different learning paradigms. This article presents the system's gen
eral characteristics and its use in designing ANN using the standard b
ackpropagation learning law.