Simple K-star Categorial Dependency Grammars and their Inference

Denis Béchet, Annie Foret
Proceedings of The 13th International Conference on Grammatical Inference, PMLR 57:3-14, 2017.

Abstract

We propose a novel subclass in the family of Categorial Dependency Grammars (CDG), based on a syntactic criterion on categorial types associated to words in the lexicon and study its learnability. This proposal relies on a linguistic principle and relates to a former non-constructive condition on iterated dependencies. We show that the projective CDG in this subclass are incrementally learnable in the limit from dependency structures. In contrast to previous proposals, our criterion is both syntactic and does not impose a (rigidity) bound on the number of categorial types associated to a word.

Cite this Paper

BibTeX
@InProceedings{pmlr-v57-bechet16, title = {Simple K-star Categorial Dependency Grammars and their Inference}, author = {Béchet, Denis and Foret, Annie}, booktitle = {Proceedings of The 13th International Conference on Grammatical Inference}, pages = {3--14}, year = {2017}, editor = {Verwer, Sicco and Zaanen, Menno van and Smetsers, Rick}, volume = {57}, series = {Proceedings of Machine Learning Research}, address = {Delft, The Netherlands}, month = {05--07 Oct}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v57/bechet16.pdf}, url = {http://proceedings.mlr.press/v57/bechet16.html}, abstract = {We propose a novel subclass in the family of Categorial Dependency Grammars (CDG), based on a syntactic criterion on categorial types associated to words in the lexicon and study its learnability. This proposal relies on a linguistic principle and relates to a former non-constructive condition on iterated dependencies. We show that the projective CDG in this subclass are incrementally learnable in the limit from dependency structures. In contrast to previous proposals, our criterion is both syntactic and does not impose a (rigidity) bound on the number of categorial types associated to a word.} }
Endnote
%0 Conference Paper %T Simple K-star Categorial Dependency Grammars and their Inference %A Denis Béchet %A Annie Foret %B Proceedings of The 13th International Conference on Grammatical Inference %C Proceedings of Machine Learning Research %D 2017 %E Sicco Verwer %E Menno van Zaanen %E Rick Smetsers %F pmlr-v57-bechet16 %I PMLR %P 3--14 %U http://proceedings.mlr.press/v57/bechet16.html %V 57 %X We propose a novel subclass in the family of Categorial Dependency Grammars (CDG), based on a syntactic criterion on categorial types associated to words in the lexicon and study its learnability. This proposal relies on a linguistic principle and relates to a former non-constructive condition on iterated dependencies. We show that the projective CDG in this subclass are incrementally learnable in the limit from dependency structures. In contrast to previous proposals, our criterion is both syntactic and does not impose a (rigidity) bound on the number of categorial types associated to a word.
RIS
TY - CPAPER TI - Simple K-star Categorial Dependency Grammars and their Inference AU - Denis Béchet AU - Annie Foret BT - Proceedings of The 13th International Conference on Grammatical Inference DA - 2017/01/16 ED - Sicco Verwer ED - Menno van Zaanen ED - Rick Smetsers ID - pmlr-v57-bechet16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 57 SP - 3 EP - 14 L1 - http://proceedings.mlr.press/v57/bechet16.pdf UR - http://proceedings.mlr.press/v57/bechet16.html AB - We propose a novel subclass in the family of Categorial Dependency Grammars (CDG), based on a syntactic criterion on categorial types associated to words in the lexicon and study its learnability. This proposal relies on a linguistic principle and relates to a former non-constructive condition on iterated dependencies. We show that the projective CDG in this subclass are incrementally learnable in the limit from dependency structures. In contrast to previous proposals, our criterion is both syntactic and does not impose a (rigidity) bound on the number of categorial types associated to a word. ER -
APA
Béchet, D. & Foret, A.. (2017). Simple K-star Categorial Dependency Grammars and their Inference. Proceedings of The 13th International Conference on Grammatical Inference, in Proceedings of Machine Learning Research 57:3-14 Available from http://proceedings.mlr.press/v57/bechet16.html.

Related Material