Introduction
Some of the discussion of how these graphs are created has been moved down to a methodology section.
The bot was offline for a two intervals of a few days each in September. This shows up as gaps in the otherwise-not-very-informative scores and time section.
Time delay
This graph compares the delay between when a toot is created and when the bot boosts it. As last time, higher scoring toots tend to be boosted sooner after their initial creation – which makes sense considering how the bot scores toots and considers which to boost. This month I have dropped a linear regression onto the graph which coincides with the impression that higher scoring toots are boosted more quickly, but with quite a limited correlation.
Scores and time
The bot assigns a score to each toot it sees, then boosts the toots which get the highest score. This can mean that when there aren’t many toots, toots with lower scores might get boosted which otherwise wouldn’t be amplified.
I was about to drop the old scatter graphs but have kept them - they show what a range of accounts are boosted, something of the range of scores assigned, but if you’ve read this analysis before you may want to skip to the day of week variation, below.
But even when I crop out the outliers, there’s not much visible pattern to see, except fewer toots boosted on weekends.
Accounts boosted
Anyone who follows the bot and will see that it tends to boost some accounts a lot. This isn’t very surprising since (a) it doesn’t follow many accounts and (b) even those accounts who do post about ADN topics don’t all consistently use the hashtags which the bot listens for.
Still, a histogram shows that nknews, JPZanders and Livableworld are most likely to score well and be boosted. Jean-Pascal Zanders (@JPZanders@mastodon.world) has moved into second position. There is considerable overlap between Council for a Livable World (@Livableworld@mastodon.world) and Nukes of Hazard (@nukes_of_hazard@mastodon.world) but I have not tried to filter or de-duplicate these.
Scores of toots from regularly boosted accounts
Looking at the scores of toots from the most frequently boosted accounts, there is some variation. The algorithm is very simple and rewards use of hashtags and pays attention to likes and boosts from other users.
Day of week variation
There always seemed to be some weekly cycle in the scores of the toots boosted. Adjusted for the days of the week in the month, but not for the downtime, this gives us a slow weekend and an end-of-the-work-week rush.
Scoring by day of week
That was a very boring monochrome chart, but if we break all the boosted toots into four quartile boxes from lowest to highest quartile by score, and then look at which quartiles show up across which weekdays, we get an interesting picture (as in July and August). These charts are made using Bob Rudis’ (@hrbrmstr@mastodon.social) waffle library.
This chart isn’t just colourful, it shows that Thursday and Friday have roughly comparable numbers of boosts. This time quality is better relatively on Wednesday and on Thursday.
Sites referred to
Looking at which sites the toots link to there is greater variation, though www.nknews.org, gets 40% of the links. Beyond that, there is a mix of press, NGOs, and other specialised media outlets.
Once you can make waffles, you like to make waffles.
Keyword frequency and topic analysis
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, and where they may overlap.
More than half (52%) all the toots referred to nuclear, followed by Korea (20%) , “missile” (17%) and Russia (15%). This month, “Korea” and “missile” together (14%) came slightly less frequently than “Korea” and “nuclear” (15%). 41 of the boosted toots (out of 202 total – 20%) didn’t fall into any category.
“nuclear” and “power” appear together in 5 toots. The bot may be amplifying toots about nuclear power plants, but nuclear weapons are regularly discussed as tools of state power.
Comments
Last month I had ambitions of doing a six month retrospective. I think it will have to be a seven month retrospective.
Methodology
This is a rough post-facto analysis of the behaviour of the ICYMI (ADN) bot. The bot spits out some rudimentary logs as it works (in fact as part of how it processes creating a list of toots to boost, and tracking which of those it has boosted and which it still has to boost) and this stores high-level data about the promoted toots (but not the discarded ones). I use the logs along with data drawn from the server using the rtoot
library to slice and dice the data and try to present the data graphically. This is still largely an exploratory rather than explanatory exercise.