Published online by Cambridge University Press: 15 March 2005
In this paper we investigate how it is possible to recover anautomaton from a rational expression that has been computed from thatautomaton. The notion of derived term of an expression, introduced by Antimirov,appears to be instrumental in this problem. The second important ingredient is the co-minimization of anautomaton, a dual and generalized Moore algorithm on non-deterministicautomata. We show here that if an automaton is then sufficiently “decorated”, thecombination of these two algorithms gives the desired result. Reducing the amount of “decoration” is still the object of ongoing investigation.