This paper describes a new technique for obtaining
measures of semantic relatedness. Like other recent
approaches, it uses Wikipedia to provide structured world
knowledge about the terms of interest. Our approach is
unique in that it does so using the hyperlink structure of
Wikipedia rather than its category hierarchy or textual
content. Evaluation with manually defined measures of
semantic relatedness reveals this to be an effective
compromise between the ease of computation of the former
approach and the accuracy of the latter.
Description
CiteULike: An effective, low-cost measure of semantic relatedness obtained from Wikipedia links
Authors propose link-based method for similarity computation in Wikipedia. It is based on the intersection and number of incoming links for two articles. Measure shows good performance for two similarity benchmarks. Resolving ambiguity is also discussed. It is based on anchor links which results into up to 26 candidates per article. The following approaches are used: most common, most related, highest (common+related), sequential and anchor-based (check for two words appearing in anchor).
%0 Conference Paper
%1 AAAI:2008:milne
%A Milne, David
%A Witten, Ian H.
%B AAAI 2008
%D 2008
%K toread wikisempaths
%T An effective, low-cost measure of semantic relatedness obtained from Wikipedia links
%U http://www.aaai.org/Papers/Workshops/2008/WS-08-15/WS08-15-005.pdf
%X This paper describes a new technique for obtaining
measures of semantic relatedness. Like other recent
approaches, it uses Wikipedia to provide structured world
knowledge about the terms of interest. Our approach is
unique in that it does so using the hyperlink structure of
Wikipedia rather than its category hierarchy or textual
content. Evaluation with manually defined measures of
semantic relatedness reveals this to be an effective
compromise between the ease of computation of the former
approach and the accuracy of the latter.
@inproceedings{AAAI:2008:milne,
abstract = {This paper describes a new technique for obtaining
measures of semantic relatedness. Like other recent
approaches, it uses Wikipedia to provide structured world
knowledge about the terms of interest. Our approach is
unique in that it does so using the hyperlink structure of
Wikipedia rather than its category hierarchy or textual
content. Evaluation with manually defined measures of
semantic relatedness reveals this to be an effective
compromise between the ease of computation of the former
approach and the accuracy of the latter.},
added-at = {2012-09-19T15:38:39.000+0200},
author = {Milne, David and Witten, Ian H.},
biburl = {https://www.bibsonomy.org/bibtex/2ba9ad8be8431e2a5cb4f9e2f7db7f5ce/psinger},
booktitle = {AAAI 2008},
citeulike-article-id = {4956624},
citeulike-linkout-0 = {http://www.aaai.org/Papers/Workshops/2008/WS-08-15/WS08-15-005.pdf},
comment = {Authors propose link-based method for similarity computation in Wikipedia. It is based on the intersection and number of incoming links for two articles. Measure shows good performance for two similarity benchmarks. Resolving ambiguity is also discussed. It is based on anchor links which results into up to 26 candidates per article. The following approaches are used: most common, most related, highest (common+related), sequential and anchor-based (check for two words appearing in anchor).},
description = {CiteULike: An effective, low-cost measure of semantic relatedness obtained from Wikipedia links},
interhash = {f8b0b3ba8f4a1c20e3d5d732a221f102},
intrahash = {ba9ad8be8431e2a5cb4f9e2f7db7f5ce},
keywords = {toread wikisempaths},
posted-at = {2010-04-20 11:06:23},
priority = {0},
timestamp = {2012-09-19T15:38:39.000+0200},
title = {An effective, low-cost measure of semantic relatedness obtained from Wikipedia links},
url = {http://www.aaai.org/Papers/Workshops/2008/WS-08-15/WS08-15-005.pdf},
year = 2008
}