[Corpora-List] associative experiments, comparative linguistics

[Justin Washtell]

Hello Lev,

Past experiments in gathering associations from humans tend to be found in the phychology literature and some are quite old. See Washtell & Markert (2009) for an overview of some existing datasets gathered in this way. Also take a look at www.wordassociation.org. See various works by Reinhard Rapp and by Yves Peirsman for investigations in German and Dutch respectively (although, off the top of my head, I'm not sure what to recommend with regard to comparative analyses across languages). With respect to computational association measures the above authors will get you started, but also visit Stefan Evert's website at www.collocations.de for an excellent and quite comprehensive presentation of various measures. With respect to terminology, the terms "(semantic) relatedness" and "(semantic) similarity" are both often used interchangeably to imply lexical association, however in some works (and arguably more correctly) "similarity" relates specifically to higher-order or paradigmatic associations, as opposed to first-order or syntagmatic associations. The works mentioned mostly deal with predicting human associations from the distributional characteristics of words in a corpus, but there also exist measures based on Wordnet (Ted Pedersen, 2004), and on dependency relations in parsed text (e.g. Pado & Lapata, 2007).

Hope that helps!

Justin Washtell
University of Leeds

________________________________________
From: corpora-bounces at uib.no [corpora-bounces at uib.no] On Behalf Of Lev Kundin [lev.kundin at gmail.com]
Sent: 17 August 2009 23:06
To: corpora at uib.no
Subject: [Corpora-List] associative experiments, comparative linguistics

Dear all,

I would appreciate if someone could point me into the direction of
some research/experiments(off/on-line)/articles/any other materials in
the area of gathering word associations, various methods of analysing
them and/or comparative analysis of word associations among different
languages. Any help will be highly appreciated, including corrections
on the terminology of this research area that I have hopefully vaguely
identified.

Thank you very much in advance.

Best regards,
Lev.

_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora

_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora