HOME   ::  Conference List   ::   Conference Paper
Chang, N. C. and Maia, T. V. (2001) Grounded Learning of Grammatical Constructions. In 2001 AAAI Spring Symposium on Learning Grounded Representations.
Bookmark:  

Full-text
   URL: http://www.icsi.berkeley.edu/NTL/papers/NChang01.pdf
   Cached: PDF-111K   
   SAVE AS an easy-to-recall long filename:
      Filename format: author--year--title   PDF-111K   
      Filename format: author--year--title--journal|proceedings|...--pages   PDF-111K   

Related links
   CiteSeer: http://citeseer.ist.psu.edu/chang01grounded.html
  Web search: Google Web Search   ::   Google Scholar
  Within this site: Cited by (3)    References (16)

Paper at a Glance 

Grounded Learning of Grammatical Constructions
Nancy C. Chang # and Tiago V. Maia y
International Computer Science Institute
1947 Center Street, Suite 600, Berkeley, CA 94704
fnchang, maiag@icsi.berkeley.edu
Abstract We describe a model of grammar learning in which all linguistic units are grounded in rich conceptual rep­ resentations, and larger grammatical constructions in­ volve relational mappings between form and meaning that are built up from smaller (e.g., lexical) construc­ tions. The algorithm we describe for acquiring these grammatical constructions consists of three separate but interacting processes: an analysis procedure that uses the current set of constructions to identify mappings be­ tween an utterance and its accompanying situation; a hypothesis procedure that creates new constructions to account for remaining correlations between these two domains; and reorganization processes that generalize existing constructions on the basis of similarity and co­ occurrence. The algorithm is thus grounded not only in the twin poles of form and meaning but also, more im­ portantly, in the relational mappings between the two.
Introduction Language is often branded as a prime example of the hu­ man capacity for abstract symbol manipulation, with formal approaches to syntax in the forefront. Treating units of lan­ guage (typically words or parts of speech) as autonomous and disembodied -- that is, disregarding any conceptual ba­ sis or meaning with which they may be associated -- has yielded notable success in speech recognition and parsing technology. As yet, however, this approach has made com­ paratively little headway toward effective natural language understanding and learning systems. This disparity is readily explained within the cognitive linguistic tradition, which views linguistic knowledge at all levels as consisting of mappings between the domains of form and meaning (Langacker 1991), where form typically refers to the speech or text stream and meaning ...
BibTex
@inproceedings{chang01groundedLearning,
  author={N. C. Chang and T. V. Maia},
  title={Grounded Learning of Grammatical Constructions},
  year={2001},
  booktitle={2001 AAAI Spring Symposium on Learning Grounded Representations},
  url={http://www.isrl.uiuc.edu/~amag/langev/paper/chang01groundedLearning.html}
}

 HOME   ::  Conference List   ::   Conference Paper Comments to: junwang4 you-know-at gmail.com Last update: 2/2/08