We study the performance of two representations of word meaning in learning noun-modifier semantic relations. One representation is based on lexical resources, in particular WordNet, the other on a corpus. We experimented with decision trees, instance-based learning and Support Vector Machines. All these methods work well in this learning task. We report high precision, recall and F-score, and small variation in performance across several 10-fold cross-validation runs. The corpus-based method has the advantage of working with data without word-sense annotations and performs well over the baseline. The WordNet-based method, requiring wordsense annotated data, has higher precision.
%0 Conference Paper
%1 Nastase:EtAl:06
%A Nastase, Vivi
%A Shirabad, Jelber Sayyad
%A Sokolova, Marina
%A Szpakowicz, Stan
%B Proceedings of the 21st National Conference on Artificial Intelligence
%D 2006
%K 2006 aaai compounds
%T Learning Noun-Modifier Semantic Relations with Corpus-based and WordNet-based Features
%U http://www-etud.iro.umontreal.ca/~sokolovm/compare_contexts_NMRs.pdf
%X We study the performance of two representations of word meaning in learning noun-modifier semantic relations. One representation is based on lexical resources, in particular WordNet, the other on a corpus. We experimented with decision trees, instance-based learning and Support Vector Machines. All these methods work well in this learning task. We report high precision, recall and F-score, and small variation in performance across several 10-fold cross-validation runs. The corpus-based method has the advantage of working with data without word-sense annotations and performs well over the baseline. The WordNet-based method, requiring wordsense annotated data, has higher precision.
@inproceedings{Nastase:EtAl:06,
abstract = {We study the performance of two representations of word meaning in learning noun-modifier semantic relations. One representation is based on lexical resources, in particular WordNet, the other on a corpus. We experimented with decision trees, instance-based learning and Support Vector Machines. All these methods work well in this learning task. We report high precision, recall and F-score, and small variation in performance across several 10-fold cross-validation runs. The corpus-based method has the advantage of working with data without word-sense annotations and performs well over the baseline. The WordNet-based method, requiring wordsense annotated data, has higher precision.},
added-at = {2007-05-18T01:56:52.000+0200},
author = {Nastase, Vivi and Shirabad, Jelber Sayyad and Sokolova, Marina and Szpakowicz, Stan},
biburl = {https://www.bibsonomy.org/bibtex/2d2212335ad1b6dafc2f33c7db3caa6c8/seandalai},
booktitle = {Proceedings of the 21st National Conference on Artificial Intelligence},
interhash = {eeb89a47d3bf0dbde4f44ea35dc4cd5e},
intrahash = {d2212335ad1b6dafc2f33c7db3caa6c8},
keywords = {2006 aaai compounds},
timestamp = {2007-05-18T01:56:52.000+0200},
title = {Learning Noun-Modifier Semantic Relations with Corpus-based and WordNet-based Features},
url = {http://www-etud.iro.umontreal.ca/~sokolovm/compare_contexts_NMRs.pdf},
year = 2006
}