Abstract
A new general method for inference of regular languages using nondeterministic automata as output has recently been developed and proved to converge. The aim of this paper is to describe and analyze the behavior of two implementations of that method and to compare it with two well known algorithms for the same task. A complete set of experiments has been carried out and the results of the new algorithms improve the existing ones both in recognition rates as in sizes of the output automata. © 2008 Springer-Verlag Berlin Heidelberg.
Cite
CITATION STYLE
García, P., De Parga, M. V., Álvarez, G. I., & Ruiz, J. (2008). Learning regular languages using nondeterministic finite automata. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5148 LNCS, pp. 92–101). https://doi.org/10.1007/978-3-540-70844-5_10
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
