A genetic algorithm for grammatical inference
In this paper, genetic algorithms have been shown to be a useful tool for the induction of context-free grammars from positive and negative examples of a language. Grammars for the language of correct parentheses expressions, for the language of equal numbers of a's and b's, for several regular languages, and for a small natural language subset have been inferred more or less correctly. Further experimentation will have to show whether this technique is applicable to more complex languages.
- 1.K.S. Fu and T.L. Booth. Grammatical inference: Introduction and survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 8:343–375, 1986.Google Scholar
- 2.J.H. Holland. Adaptation in Natural and Artificial Systems. University of Michigan Press, Ann Arbor, 1975.Google Scholar
- 3.S. Sen and J. Janakiraman. Learning to construct pushdown automata for accepting deterministic context-free languages. In G. Biswas, editor, SPIE Vol. 1707: Applications of Artificial Intelligence X: Knowledge-Based Systems, pages 207–213. 1992.Google Scholar
- 4.P. Wyard. Context free grammar induction using genetic algorithms. In R.Belew and L.B.Booker, editors, Proceedings of the Fourth Conference on Genetic Algorithms ICGA'92. Morgan Kaufmann, 1992.Google Scholar
- 5.H. Zhou and J.J. Grefenstette. Induction of finite automata by genetic algorithms. In Proceedings of the 1986 IEEE International Conference on Systems, Man and Cybernetics, pages 170–174, Atlanta, GA, 1986.Google Scholar