Learning deterministic regular grammars from stochastic samples in polynomial time

• Published in: RAIRO. Informatique théorique et applications Vol. 33; no. 1; pp. 1 - 19
• Main Authors: Carrasco, Rafael C., Oncina, Jose
• Format: Journal Article
• Language: English
• Published: Paris EDP Sciences 01.01.1999
• Subjects: Applied sciences; Automata. Abstract machines. Turing machines; Computer science; control theory; systems; Exact sciences and technology; Language theory and syntactical analysis; Theoretical computing; Formal language; Finite automaton; Identification; Grammar; Stochastic language; Stochastic grammar; Algorithm; Convergence; Learning; Regular grammar; Computer program; Stochastic deterministic regular language; Stochastic automaton; Regularity; Deterministic language
• ISSN: 0988-3754; 1290-385X
• DOI: 10.1051/ita:1999102

In this paper, the identification of stochastic regular languages is addressed. For this purpose, we propose a class of algorithms which allow for the identification of the structure of the minimal stochastic automaton generating the language. It is shown that the time needed grows only linearly with the size of the sample set and a measure of the complexity of the task is provided. Experimentally, our implementation proves very fast for application purposes. Dans cet article, on étudie l'identification de langages réguliers stochastiques. Dans ce but, nous proposons une classe d'algorithmes permettant l'identification de la structure de l'automate stochastique minimal qu'engendre le langage. On trouve que le temps nécessaire croît linéairement avec la taille de l'échantillon et on donne une mesure de la complexité de l'identification. Expérimentalement, notre mise en œuvre est très rapide, ce qui la rend très intéressante pour des applications.