[Lingtyp] tools for calculating morpheme to word ratios

[Matías Guzmán Naranjo]

Dear Alexander,

the answer will depend a lot on the type of data you have and what you want
exactly.
We describe two generic alignment tools here
https://scholarworks.umass.edu/scil/vol4/iss1/21/ , but these don't find
'morphemes'.
That, you'd have to extract from the alignments.
Alternatively, you can try morfessor:
https://morfessor.readthedocs.io/en/latest/, but results will depend on
many factors, like corpus size and variety.
One thing though, Quechua is relatively 'easy' and you could try to do
'morpheme' segmentation by hand, just by listing everything you consider to
be a morpheme, and checking each word.

But this might be an XY problem (https://en.wikipedia.org/wiki/XY_problem),
so maybe you want to tell us exactly what is it you want to do?

Best,

El mié, 25 oct 2023 a las 23:46, Alexander Rice (<ax.h.rice at gmail.com>)
escribió:

> Howdy folks
>
> I'm interested in finding average morphemes per word in a Quechuan
> language I'm working on. I'd be looking at overall average morphemes per
> word, nouns vs. verbs, etc.
>
> I'm wondering if there are software tools or R scripts out there for doing
> this kind of thing. I could do it myself from scratch in R, but why
> reinvent the wheel if you don't need to?
>
> Any links or references to resources/guides, pointers, advice, etc. is
> most appreciated.
>
> best,
> --Alex
>
> --
> Alexander Rice, (he, him, his)
> <https://www.su.ualberta.ca/services/thelanding/learn/pronouns/>, PhD
> Candidate
> Department of Linguistics, University of Alberta
> 3-27 Assiniboia Hall
> https://sites.google.com/view/arice
>
>
>
> <#m_7609474339975931584_m_221812164836572711_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> https://listserv.linguistlist.org/cgi-bin/mailman/listinfo/lingtyp
>


-- 
Dr. Matías Guzmán Naranjo
Sprachwissenschaftliches Seminar
Albert-Ludwigs-Universität Freiburg
https://mguzmann89.gitlab.io/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20231026/b74f1231/attachment.htm>