[Corpora-List] New semantic vectors available for English and Dutch

Marc Brysbaert Marc.Brysbaert at UGent.be
Thu Apr 21 10:33:44 CEST 2016


Dear all,

Research has profited tremendously from the LSA semantic vectors and the fact that they were easy to calculate on this website<http://lsa.colorado.edu/>. In the meantime, significant progress has been made in the quality of the vectors, which unfortunately has been more difficult to access for many language researchers. In a review paper, Mandera et al.<http://crr.ugent.be/papers/Mandera_et_al_JML_2016.pdf> (JML, in press) evaluated the various vectors available and saw big differences in quality as a function of the algorithm used and the corpus on which the vectors were calculated.

More importantly, a website<http://zipf.ugent.be/snaut/> has been created that allows you to calculate the semantic distance between any two words in English and in Dutch. There are two versions: one web version with lemmas only and one standalone program with all words. Please download the standalone program if you want to do extensive searches, as otherwise the site might crash.

Enjoy!

Marc

Marc Brysbaert Department of Experimental Psychology Ghent University Henri Dunantlaan 2 B-9000 Gent Belgium Tel. +32 9 264 94 25 Fax. +32 9 264 64 96 E-mal: marc.brysbaert at ugent.be<mailto:marc.brysbaert at ugent.be> Website: http://crr.ugent.be/members/marc-brysbaert

-------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 4127 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20160421/06251fad/attachment.txt>



More information about the Corpora mailing list