An inductive method with genetic algorithm for learning phrase-structure-rule of natural language
- 21 Downloads
This paper describes an Inductive method with genetic search which learns attribute based phraserule of natural language from set of preclassified examples. Every example is described with some attributes/values. This algorithm takes an example as a seed, generalizes it by genetic process, and makes it cover as many examples as possible. We use genetic operator in population to perform a probabilistic parallel search in rule space and it will reduce greatly possible rule search space compared with many other inductive methods. In this paper, we give description of attribute, word, dictionary and rule at first; then we describe learning algorithm and genetic search process, and at last, we give a computing method about quility of rule C(r).
Key wordsPhrase-rule Example Generalization Induction Genetic Algorithm
Unable to display preview. Download preview PDF.
- Gabriec Fiolet al. “A New Perspective in the Inductive Acquisition of Knowledge from example”. Machine Learning: ECML'93, Pavel B. Brazdii (Ed.) Springer Verlag 1993.Google Scholar
- Valiant, L.G., “A Theory of the Learnable”. Communication of ACM. Vol.27., No.11, November 1984.Google Scholar
- Gilles Venturini. “SIA: A Supervised Inductive Algorithm with Genetic search for learning attributes based concepts “Machine Learning: ECML'93, Pavel B. Brazdil (Ed.) Springer Verlag 1993.Google Scholar
- John T. Maxwell III and Ronald M. Kaplan. “The Interface Between Phrasal and Functional Constraints”, Computational Linguistics, Vol. 19, No. 4. December 1993.Google Scholar
- Yasubumi, “Learning Context-free Grammars From Structural Data in polynomial time”, Theoretical Computer Science 76 (1990). North-Holland.Google Scholar
- Chen Zhao Xiong. Advance on Machine Translating. Electrical Industry Press, (China) 1993Google Scholar
- Chen Zhong Ming, Introduction to Chinese logic, People Press. 1993.Google Scholar