1/ When the Russian invasion of Ukraine started, we started collecting tweets in order to be able to identify propaganda. The (very long) thread below covers the story of #istandwithrussia and #istandwithputin. Inspired by @marcowenjones' work.
2/ First about the data: we collected ~210k tweets for #istandwithputin, ~120k tweets for #istandwithrussia, plus about 20k attached images. There is significant overlap between the two hashtags, but there are also differences.
3/ I'm going to try to answer the "when, who and what" for these two hashtags. All results are preliminary, still some work planned.
4/ The most intense activity happened on the 2nd and 3rd of March, with a peak of over 100k tweets for #istandwithputin and over 50k tweets for #istandwithrussia. One day before Twitter was banned in Russia.
5/ First, let's look at the accounts interacting, starting with #istandwithputin. This is the mentions network, including (some) retweets for the tweets we collected. Still working on the retweet network itself.
6/ Here are the top 10 tweets for #istandwithputin from our data. About half of the accounts started tweeting in early 2022, have very few followers, yet were apparently very successful.
7/ The lonely top tweet against Putin belongs to @NickKnudsenUS, but there was a significant effort to spread anti-Putin messages on these hashtags from other users as well (@AndrewOnSeeAIR was another significant account).
8/ We know from media reports that the hashtag was most popular in India and our data confirms that. More details below on the actual content.
9/ The mentions network for #istandwithrussia is less dominated by Indian users, with much more African users participating.
10/ A notable participant in the conversation is Duduzile Zuma-Sambudla, daughter of former South African president Zuma. A lot more anti-Putin voices are visible, with @KylaInTheBurgh and @AndrewOnSeeAIR (again) being the most mentioned.
11/ So what are the topics discussed? We extracted named entities using an AI text processor (hello @spacy_io), created co-occurrence networks and here is what we found:
12/ For #istandwithputin the entities strongly point to - surprise - a lot of whatabout-ism. All NATO/US wars are referenced, plus a lot of India-related content. More details below.
13/ For #istandwithrussia it's more of the same, but with more references to Africa.
14/ We identified 6 big topics in the combined tweets: Russia is good, Ukrainians are nazis, the West is bad, the West is bad to Africa, India and Russia are best friends, plus assorted general conspiracy content.
15/ Let's start with "Russia is good". These two images are, by far, the most prevalent. I'm guessing the two bear cubs are the Donetsk and Luhansk republics?
16/ A lot of V-s and Z-s with a strong Call of Duty vibe.
17/ Even more Z-s, but in color.
18/ Lots of Putin & the elements, the Putin in the rain image was very popular.
19/ Last but not least, a lot of Russia fighting bravely agains its enemies.
20/ The "Ukrainians are nazis" topic is very repetitive, with the same images of the Azov battalion being reshared.
21/ "The West is bad" is the main topic, at least media-wise.
22/ So many lists of what the West did.
23/ Palestine is a significant topic, probably the most memes for any country.
24/ "The West is bad to Africa" has a lot of Gaddafi, enough for it to deserve its own collage.
25/ Lots of maps referencing Africa's colonial past and European exploitation.
26/ Lots of references to Western racism towards Africa.
27/ Russia is presented as a good friend of Africa, with an emphasis on Putin's years in Tanzania and his current good relations with African leaders.
28/ "India and Russia are best friends" is composed primarily of photos of Putin and Modi, plus references to "akhand bharat" (Undivided India) and "akhand USSR" (i'm guessing Undivided USSR).
29/ The evacuation of Indian students from Ukraine also featured prominently.
30/ Historical & international relations references are also plentiful.
31/ To conclude, there's plenty of the conspiracy tropes we all know and love.
32/ If you've reached this far, thank you, hope this was informative :).
• • •
Missing some Tweet in this thread? You can try to
force a refresh
🧵 1/ Here's a more in depth thread about the actors in the Russian propaganda reaction to the Bucha massacre. We collected & analysed 4.2M tweets & retweets between April 3-8, so this should be quite representative of the overall conversation.
2/ You can compare this with my original reaction thread on the same topic, the preliminary results stand, but this adds a lot of detail.
3/ On April 3 we identified a Spanish-language cluster of pro-Russia accounts. For now it is small and marginal, but will become significantly bigger in the coming days.
🧵 1/ New thread about the Russian propaganda reaction to the Bucha massacre. We collected 193k related tweets & retweets on the April 3rd and and 295k on the 4th. Usual caveats apply, results are preliminary, read on for more.
2/ About 300k tweets are in English, with Spanish as the second most popular language (75k) and Italian in third place, with 30k tweets. On the mentions network for the 3rd there is a clearly visible propaganda cluster (mostly in spanish) at the top of the chart.
3/ The same cluster is visible in the mentions network the 4th, with a combination of Russian gov, state media accounts & other spreaders. The Russian Embassy in Spain played a significant role, as did RT reporter Helena Villar.
🧵1/ This is a new thread that covers the evolution of the "Pentagon secret biolabs in Ukraine" conspiracy. It covers content for March 3-20 and includes things from George Soros to Russian disinformation assets in action, with a healthy dose of US & Chinese participation.
2/ First about the data: we are currently collecting as many tweets as we can that contain the words Ukraine or Russia. Hundreds of thousands per day, much more in the beginning. Data referenced in this thread is just a subset of a (much) larger dataset.
3/ We selected ~158k tweets that contain any variants of "bioweapons", "biolabs", "biowarfare" etc. Most tweets are in English (~150k), but there is a significant number of tweets in German, hence the "biowaffen" that you'll see in a lot of charts.