[Air-L] About estimation of quantity of Tweets per keyword

[Shulman, Stu]

Yes, if the date range is the last 12 months. We can get you up to 20,000
Tweets for a pilot test via DiscoverText.

Anyone with a valid academic email can book a free meeting for free
training to get the free data sample:

https://calendly.com/discovertext

Just to reiterate, on the eve of a consequential election in the United
States, we can get custom Twitter datasets and teach you how to parse them
in about 30-45 minutes. There is no weeklong, or monthlong, or semester
requirement to get to a meaningful finding or "ah hah" moment. To my
knowledge, DiscoverText is the only free scientific platform that (i)
embeds the Twitter display, (ii) enables crowdsource annotation, and
(iii-vi) provides tools for measurement of annotator reliability,
adjudication of disagreements, and production of gold standard training
sets for an internal machine-learning capability. It does other cool stuff
as well. If you need datasets larger than 20,000 items, there is a
corresponding data access and hourly fee. You will learn things about the
data in the first 5 minutes and they will not be in the form of a word
cloud.

There is also now TrustDefender.net, which is DiscoverText with some
upgrades specifically for teaching. One is, we have simplified the
onboarding of entire classes for collaborative experiments. It's a few
clicks and everyone gets a license and when they are registered and logged
in they can collaborate immediately with any other class member of the
professor via a "peer" network first launched as open source web-based
software in 2007 at the University of Pittsburgh. As we roll through year
seventeen providing access to tools, methods, and novel analytical theories
to academics, I invite you to join the more than 1,000 scholars who have
published as a result of using our software.

As I note in my first paper written since 2009: "Elon Musk likely prefers
academics to move on. There is a popular but misleading notion now that
only extremists, particularly right-wing fans of Tesla, Trump, Putin,
Russia, and SpaceX, dominate the platform in North America. This is
demonstrably false. A massive and diverse network of identity, hashtag, and
political resistance communities is constantly organizing, sharing,
strategizing, and taking political action to get out voters, raise
awareness, and trend liberal, progressive, and pro-democracy themes on
Twitter. Journalists also use Twitter as a real time content filter shaping
their attention cycle when deciding what to cover. Academics use Twitter to
network and promote their work. Any serious candidate for political office
is on Twitter. Academic researchers must access, work with, and report on
this data."

~Stu



On Mon, Oct 7, 2024 at 7:54 PM Xanat Meza via Air-L <air-l at listserv.aoir.org>
wrote:

> Hello everyone,
> I know that this question might not have an answer given the changes
> Twitter has gone through in the last year, but I thought it was worth a
> shot.
> Is there any online tool that could estimate the quantity of Tweets
> containing a keyword within a specific time frame?
> Regards,
> Xanat V. Meza
>
> Ph.D. Kansei, Behavioral and Brain SciencesUniversity of Tsukuba
> M.A. Media and Communication
> Yeungnam University
> B.D. Graphic Communication Design
> Universidad Autonoma Metropolitana
> _______________________________________________
> The Air-L at listserv.aoir.org mailing list
> is provided by the Association of Internet Researchers http://aoir.org
> Subscribe, change options or unsubscribe at:
> http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
>
> Join the Association of Internet Researchers:
> http://www.aoir.org/
>


-- 
Dr. Stuart W. Shulman
Founder and CEO, Texifter
Editor Emeritus, *Journal of Information Technology & Politics*
ResearchGate Profile <https://www.researchgate.net/profile/Stuart-Shulman>