Have you looked at slurk from the Uni Potsdam group? https://github.com/clp-research/slurk

I'm currently collecting data for Scottish Gaelic using a modified version of one of their tasks to get two humans to interact in an information seeking/answering paradigm.

Peace,
Dave
----
David M. Howcroft
https://www.davehowcroft.com


On Wed, Jul 6, 2022 at 10:49 AM Valentin Werner <valentin.werner@uni-bamberg.de> wrote:
Dear Corpora-List members,

I was wondering whether anyone of you has some experience with using
chatbots for collecting production data in a (pseudo) social media context.

The concrete scenario I have in mind is the simulation of a text message
chat (as in WhatsApp, for instance) with a human informant interacting
with the chatbot during an exchange on a set topic (e.g. the first line
of the chatbot could be something like: "what are your plans for this
afternoon?"). The idea is to elicit informal digital conversation.

In an ideal case, there would be the opportunity to store the production
data by the human informant in a format that can be used for subsequent
analysis with standard corpus software (such as AntConc, WordSmith or
SketchEngine). I'm aware that there are several (also partly free)
chatbots available for commercial purposes (as presented on
chatbots.org, for example) but they don't really seem to allow data
export in the way described above.

Any pointers would be highly appreciated!

Valentin

--
PD Dr. Valentin Werner
Otto-Friedrich-Universität Bamberg
Englische Sprachwissenschaft
D-96045 Bamberg


_______________________________________________
Corpora mailing list -- corpora@list.elra.info
https://list.elra.info/mailman3/postorius/lists/corpora.list.elra.info/
To unsubscribe send an email to corpora-leave@list.elra.info