Elsevier

Theoretical Computer Science

Volume 123, Issue 2, 31 January 1994, Pages 273-289
Theoretical Computer Science

The word problem for inverse monoids presented by one idempotent relator

https://doi.org/10.1016/0304-3975(92)00063-WGet rights and content
Under an Elsevier user license
open archive

Abstract

We study inverse monoids presented by a finite set of generators and one relation e = 1, where e is a word representing an idempotent in the free inverse monoid, and 1 is the empty word. We show that (1) the word problem is solvable by a polynomial-time algorithm; (2) every congruence class (in the free monoid) with respect to such a presentation is a deterministic context-free language. Such congruence classes can be viewed as generalizations of parenthesis languages; and (3) the word problem is solvable by a linear-time algorithm in the more special case where e is a “positively labeled” idempotent.

Cited by (0)

Research supported by N.S.F. Grant No. DMS 8702019 and by the Center for Communication and Information Sciences, University of Nebraska, Lincoln.