Corpora: Creating wordlists / 2-5 word clusters / **freq = 1**

[Tony Berber Sardinha]

Dear Mark

You can get around this limitation in WSmith Tools by selecting your corpus
files twice, and so any clusters with a frequency of 2 will actually have a
frequency of 1.

cheers
tony.
-------------------------------------
Dr Tony Berber Sardinha
Catholic University of Sao Paulo, Brazil
tony4 at uol.com.br
www.tonyberber.f2s.com
>
> -----Mensagem Original-----
> De: Mark Davies <mdavies at ilstu.edu>
> Para: <CORPORA at HD.UIB.NO>
> Enviada em: terça-feira, 3 de abril de 2001 09:12
> Assunto: Corpora: Creating wordlists / 2-5 word clusters / **freq = 1**
>
>
> Can anyone recommend a PC-based program that creates wordlists with the
> following three characteristics:
>
> 1) 2 / 3 / 4 / 5 word clusters
> 2) ** clusters that occur as little as just one time **
> 3) wordlists of multi-million word texts (can do smaller chunks and merge
> them together)
>
> For my present needs, #2 is the most important.  I've been using WordSmith,
> and it can of course create wordlists of word clusters, but purposely
> limits the lists to only those clusters that occur two times or more.  (In
> Settings / Min/Max Frequencies / Word Frequency you can set it as low as 1,
> but for 2+ word clusters it won't actually return any clusters with a
> frequency less than 2).  This limitation does makes sense, since the number
> of clusters that occur only once will be extremely large -- easily in the
> millions of distinct strings for 4-5 word clusters.  Nevertheless, for a
> project that I am doing, this is (unfortunately) exactly what I need to do.
>
> Thanks in advance for your help.
>
> Mark Davies
>
> =======================================
> Mark Davies, Associate Professor, Spanish Linguistics
> http://mdavies.for.ilstu.edu/
>
> "Where is the wisdom we have lost in knowledge?
> Where is the knowledge we have lost in information?"
> -- T.S. Eliot
>
> 4300 Foreign Languages
> Illinois State University
> Normal, IL 61790-4300
> Voice:309/438-7975 / Fax:309/438-8038
> =======================================
>