This is a bit hackier than usual. I’ve been wanting to make wordclouds for a long time and I’ve been wanting to use wordclouds to try to look at what the ICYMI (ADN) bot boosts for a long time. I now have data for most of Q1 of 2024 (to be precise, January 08-April 03, 2024). I’ve used most of the data analysis code from previous monthly reports but done some adjusting to let me look at the text of the toots which are boosted.
Wordclouds
First, for everything that has been boosted, what are the most frequently-appearing words?
Note that I’ve a long list of words which are being ignored1 which doesn’t include “die” even though I have a number of other common German words, because it might be “die” (the English verb).
A very very basic keyword search was used to mark all boosted toots based on the content of the toot (not any linked site). This was then used to look at which topics are most frequently referred to (nuclear, biological, chemical, missile, arms control, proliferation, disarmament, Russia, power), and where they may overlap.
Once we look at the toots which haven’t been categorised into any of the other possibilities we can then see what words show up there … and perhaps it is the English verb after all.
Word clouds for nuclear toots …
and chemical toots …
are largely unsurprising.
No conclusions, no analysis
I don’t know if there is a great deal to be learned from these but it has been satisfying to do it and I’m glad I’ve figured out how to do these.
-
“The” “that” “s” “at” “with” “I” “we” “A” “on” “the” “an” “a” “and” “in” “for” “of” “to” “it” “by” “its” “about” “from” “not” “but” “our” “all” “up” “or” “over” “after” “my” “you” “his” “her” “is” “has” “have” “are” “be” “this” “as” “been” “was” “will” “said” “use” “am” “can” “der” “und” “im” “mit” “auf” “von” “zu” ↩