The paper investigates the expressive power of language equations with the operations of concatenation and symmetric difference. For equations over every finite alphabet Σ with |Σ| ≥1, it is demonstrated that the sets representable by unique solutions of such equations are exactly the recursive sets over , and the sets representable by their least (greatest) solutions are exactly the recursively enumerable sets (their complements, respectively). If |Σ| ≥ 2, the same characterization holds already for equations using symmetric difference and linear concatenation with regular constants. In both cases, the solution existence problem is Π (0,1)-complete, the existence of a unique, a least or a greatest solution is Π(0,2)-complete, while the existence of finitely many solutions is Σ(0,3)-complete.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
It is shown that equations of the form φ(X) = ψ(X), in which the unknown X is a set of natural numbers and φ, ψ use the operations of union, intersection and addition of sets S + T = {m + n | m ∈ S, n ∈T}, can simulate systems of equations φ(X1, . . . ,Xn) = φ(X1, . . . ,Xn) with 1 ≤ j ≤ l, in the sense that solutions of any such system are encoded in the solutions of the corresponding equation. This implies computational universality of least and greatest solutions of equations φ(X) = ψ(X), as well as undecidability of their basic decision problems. It is sufficient to use only singleton constants in the construction. All results equally apply to language equations over a one-letter alphabet Σ = {α}.
3
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Unambiguous conjunctive grammars with 1 nonterminal symbol are shown to be strictly weaker than the grammars with 2 nonterminal symbols, which are in turn strictly weaker that the grammars with 3 or more nonterminal symbols. This hierarchy is established by considering grammars over a one-symbol alphabet, for which it is shown that 1-nonterminal grammars describe only regular languages, 2-nonterminal grammars describe some non-regular languages, but all of them are in a certain sense sparse, whereas 3-nonterminal grammars may describe some non-regular languages of non-zero density. It is also proved that one can test a 2-nonterminal grammar for equivalence with a regular language, whereas the equivalence between a pair of 2-nonterminal grammars is undecidable.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.