Thread by @compstorylab

We have a new paper, interactive visualization, and data platform.

Nutshell: we’ve curated 100 billion tweets over 10 years to produce day-scale rank/frequency time series for n-grams in over 100 languages.

It’s a whole big thing.

A short thread—

The paper:

“Storywrangler: A massive exploratorium for sociolinguistic, cultural, socioeconomic, and political timelines using Twitter”

arxiv.org/abs/2007.12988

With storywrangler, we’re hoping to enable or enhance the computational study of any large-scale temporal phenomena where people matter including:
culture,
politics,
economics,
linguistics,
public health,
conflict,
climate change,
and
data journalism.

(For example, our own @hedonometer happiness measuring thing now runs on storywrangler.)

Here’s the storywrangler site with usage rank time series for ‘coronavirus’, ‘Black Lives Matter’, and ‘#MeToo’:

storywrangling.org/?ngrams=corona…

The site provides rank/frequency time series of 1-grams, 2-grams, and 3-grams.

The n-grams may be sequences of words, hashtags, emojis, and more.

A key feature: We account for social amplification of n-grams via retweets.

- In the paper, we show how amplification can be visualized through ‘contagiograms’.

- Online, we have time series comparisons:
storywrangling.org/details.html?n…

To possibly entrain reality in dangerous ways, we set up yet another Twitter account: @storywrangling.

So: Feel free to explore away.

We have many developments planned for storywrangler and we will appreciate your kind feedback.

Credits flash:

Team: T. Alshaabi, J. L. Adams, M. V. Arnold, J. R. Minot, D. R. Dewhurst, A. J. Reagan, C. M. Danforth, and P. S. Dodds.

Support: The Vermont Advanced Computing Center, MassMutual, Google.

Foundational support (way back when): NSF.