J. Gorman, and J. Curran. Proceedings of the ACL Workshop on Cognitive Aspects of Computational Language Acquisition, page 73--80. Prague, Czech Republic, Association for Computational Linguistics, (June 2007)
Abstract
Semantic networks have been used successfully to explain access to the mental lexicon. Topological analyses of these networks have focused on acquisition and generation. We extend this work to look at models that distinguish semantic relations. We find the scale-free properties of association networks are not found in synonymy-homonymy networks, and that this is consistent with studies of childhood acquisition of these relationships. We further find that distributional models of language acquisition display similar topological properties to these networks.
%0 Conference Paper
%1 Gorman:Curran:07
%A Gorman, James
%A Curran, James
%B Proceedings of the ACL Workshop on Cognitive Aspects of Computational Language Acquisition
%C Prague, Czech Republic
%D 2007
%I Association for Computational Linguistics
%K 2007 graphs workshop
%P 73--80
%T The Topology of Synonymy and Homonymy Networks
%U http://acl.ldc.upenn.edu/W/W07/W07-0610.pdf
%X Semantic networks have been used successfully to explain access to the mental lexicon. Topological analyses of these networks have focused on acquisition and generation. We extend this work to look at models that distinguish semantic relations. We find the scale-free properties of association networks are not found in synonymy-homonymy networks, and that this is consistent with studies of childhood acquisition of these relationships. We further find that distributional models of language acquisition display similar topological properties to these networks.
@inproceedings{Gorman:Curran:07,
abstract = {Semantic networks have been used successfully to explain access to the mental lexicon. Topological analyses of these networks have focused on acquisition and generation. We extend this work to look at models that distinguish semantic relations. We find the scale-free properties of association networks are not found in synonymy-homonymy networks, and that this is consistent with studies of childhood acquisition of these relationships. We further find that distributional models of language acquisition display similar topological properties to these networks.},
added-at = {2007-07-09T17:49:23.000+0200},
address = {Prague, Czech Republic},
author = {Gorman, James and Curran, James},
biburl = {https://www.bibsonomy.org/bibtex/25d3975c278d9737892d7065b8c52d5f4/seandalai},
booktitle = {Proceedings of the ACL Workshop on Cognitive Aspects of Computational Language Acquisition},
interhash = {58328a4beeed434893bbd26c6c6a05b4},
intrahash = {5d3975c278d9737892d7065b8c52d5f4},
keywords = {2007 graphs workshop},
month = {June},
pages = {73--80},
publisher = {Association for Computational Linguistics},
timestamp = {2007-07-09T17:49:23.000+0200},
title = {The Topology of Synonymy and Homonymy Networks},
url = {http://acl.ldc.upenn.edu/W/W07/W07-0610.pdf},
year = 2007
}