[Air-L] Custom Project Support Services - Twitter/X Data

[Stuart Shulman]

For now, we are able to support custom research projects using Twitter/X
data from the last 12 months. If you would like to learn more about our
fee-for-service offer, please book a meeting to discuss the options. We
provide DiscoverText software for free via the Web to academics and charge
hourly rates for project management and data access services. I cannot say
how long the Twitter/X data access will remain in place; it is there now
and robust.

Book a meeting here:

https://calendly.com/discovertext

With DiscoverText, you can search, filter, cluster, sample, human code and
machine classify hundreds of thousands or millions of Tweets. Teams build
reusable, transparent, shareable, crowdsource annotation models. Project
managers measure and report inter-rater reliability. Leaders or groups
adjudicate human disagreements to improve human and machine learning. This
previous feature (adjudication) is by far the most important (and least
used/understood) piece in our methods toolkit. It is the reason I got into
building software. When two or more annotators disagree, an opportunity to
learn, evolve, and grow the team and model emerges. The beneficial
byproducts of iterative annotation/adjudication cycles are noteworthy.

In DiscoverText, Tweets are displayed in the Twitter/X live display with
media previews and images embedded in the interface. Replies show the
original Tweet and the reply in the display. Deleted and suspended posts
are not displayed; we check in real time. Numerous list members have either
used or contributed to the development of the toolkit going back to 2011.
The tools were derived from a decade of NSF-funded interdisciplinary
research (2000-2010) with computer scientists, statisticians, social
scientists.

If you want to get a Twitter/X project up and running, we can help.
Twitter/X is awash in things that should be studied and better understood
inside and beyond academia. Phenomena like this:

https://discovertext.com/mentions/

If you want to write a paper lamenting how free Twitter data is gone and
only free data will suffice, I suggest talking to everyone who ever put a
serious survey in the field, ran focus groups, or did interviews, about the
nature of free versus earned data. We wrote grant proposals for all of
that. A solution is on the table; there are others. It is factually
inaccurate, though fashionable, to say the Twitter/X data is not
accessible, for now.

~Stu

---
Dr. Stuart W. Shulman
Founder and CEO, Texifter
Editor Emeritus, Journal of Information Technology & Politics