Adaptability of Lexical Acquisition for Large-scale Grammars

Kostadin Cholakov; Gertjan van Noord; Valia Kordoni; Yi Zhang

In: Ruslan Mitkov (Hrsg.). Proceedings of RANLP 2011. International Conference on Recent Advances in Natural Language Processing (RANLP-2011), September 12-14, Hissar, Bulgaria, RANLP, 9/2011.


In this paper, we demonstrate the portability of the lexical acquisition (LA) method proposed in Cholakov and van Noord (2010a). Here, LA refers to the acquisition of linguistic descriptions for words which are not listed in the lexicon of a given computational grammar, i.e., words which are unknown to this grammar. The method we discuss was originally developed for the Dutch Alpino system, and the paper shows that the method also applies to the GG (Crysmann, 2003), a computational HPSG grammar of German. The LA method obtains very similar results for German (84% F-measure on learning unknown words). Extending the GG with the lexical entries proposed by the LA method causes an important improvement in parsing accuracy for a test set of sentences containing unknown words. Furthermore, in a smaller experiment, we show that the linguistic knowledge the LA method provides can also be used for sentence generation.

ranlp2011.pdf (pdf, 69 KB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence