Partial Learning Using Link Grammars Data

Erwan Moreau (1)

1 LINA - Laboratoire d'Informatique de Nantes Atlantique

Kanazawa has shown that several non-trivial classes of cate- gorial grammars are learnable in Gold's model. We propose in this article to adapt this kind of symbolic learning to natural languages. In order to compensate the combinatorial explosion of the learning algorithm, we suppose that a small part of the grammar to be learned is given as in- put. That is why we need some initial data to test the feasibility of the approach: link grammars are closely related to categorial grammars, and we use the English lexicon which exists in this formalism.

Mots clés: Grammatical inference; Partial learning; Grammars

Domaines: Traitement du texte et du document; Apprentissage [cs.LG]

Erwan Moreau. Partial Learning Using Link Grammars Data. Grammatical Inference: Algorithms and applications. 7th International Colloquium: ICGI 2004, Oct 2004, Athens, Greece. pp.211--222. ⟨hal-00487059⟩