Problem with using EVAL and C-NNLA commands

[Jorge Ricardo Hidalgo Chagoya]

Leonid and Brian,

Thank you for your fast responses!  This are indeed terrible news. I knew 
CNNLA was not necessarily fit for Spanish as it is stated in the manual, 
but I thought my problems with EVAL were derived from defects in the 
tagging. I am currently trying to extract some variables just using the 
FREQ command. 

It would be great to have more aphasia data in AphasiaBank indeed. 
Unfortunately, the data I am working with is not my own and I still don't 
know how they plan to treat the data once the collection is finished. 

Thank you again for your replies. It is really admirable the effort you put 
in the continuous development of this project. 
Sincerely, 
Jorge Hidalgo. 
El miércoles, 6 de marzo de 2024 a la(s) 7:20:33 p.m. UTC+1, macw escribió:

> Jorge,
> Thanks for your message. I’m glad that batchalign is working okay for you 
> in terms of taggng. 
>
> Regarding C-NNLA, the origianl NNLA program from Northwestern was only 
> designed and tested fo work with English. Developing a similar instrument 
> for other languages would be a project for each language.
>
> EVAL was also never operative for Spanish. However, making it work for 
> Spanish would not be fundamentally difficult. It will just take some time. 
> Having some Spanish aphasia data n AphasiaBank would be a great moiivating 
> force getting us to make EVAL working for Spanish.
>
> — Brian MacWhinney
> Teresa Heinz Professor of Cognitive Psychology, 
> Language Technologies and Modern Languages, CMU
>
>
>
> > On Mar 6, 2024, at 7:52 AM, Jorge Ricardo Hidalgo Chagoya <
> ulysse... at gmail.com> wrote:
> > 
> > Greetings!
> > 
> > I am currently facing problems with extracting specific variables from 
> my data. Importantly, my data is in Spanish. I have previously used the 
> batchalign script to have accurate %mor and %gra tiers tags, since the 
> tagging using CLAN´s command of MOR was not as accurate. But now that I try 
> to use the EVAL and C-NNLA commands with the resulting chat files, several 
> fields appear in blank (including number of verbs and nouns). Beforehand, I 
> ran the CHECK command and corrected all detected errors. 
> > 
> > I realise that the parts of speech tags are different from one method to 
> another. My best guess is that EVAL and C-NNLA are specifically designed to 
> track specific strings in the %mor tier (for verbs: the "v" obtained by the 
> MOR command instead of the tag "verb" obtained via batchalign). 
> > 
> > My main concern is not being able to use the batchalign files. Would you 
> know if there is a way to solve this issue? Perhaps a method to modify the 
> %mor tags so they fit the EVAL and C-NNLA requirements?
> > 
> > Thank you for your attention!
> > Jorge Hidalgo. 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "chibolts" group.
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to chibolts+u... at googlegroups.com.
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/chibolts/b7a6dd92-2005-4479-ba94-50a2e1ae0372n%40googlegroups.com
> .
>
>

-- 
You received this message because you are subscribed to the Google Groups "chibolts" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/f5622821-0efd-4558-8279-c4c902c56492n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/chibolts/attachments/20240308/5baf1eac/attachment.htm>