Thread: For 2 days, I captured 105929 tweets with the hashtags #SurfExcel, #BoycottSurfExcel and #BoycottHindustanUnilever. Follow this thread for an analyse

https://twitter.com/fs0c131y/status/1105017144817999872

When I started my capture, the tweet volume was around 170 tweet per minute <=> 8300 tweets per hour

The most shared images are:
1181 pbs.twimg.com/media/D1XplRPV…
532 pbs.twimg.com/media/D1T4DugX…
505 pbs.twimg.com/media/D1TP76nW…
420 pbs.twimg.com/media/D1Yrf-fV…
404 pbs.twimg.com/media/D1ScmmLW…

The most used hashtags are:
41154 #surfexcel
13123 #boycottsurfexcel
3289 #boycotthindustanunilever
2796 #loksabhaelections2019
2486 #हिंदुओं
2106 #हिंदू
1750 #electioncommission
1424 #boycotthulproducts
1338 #indvaus
904 #bollywood
764 #holi
675 #hindustanunilever

Among the 105928 tweets I captured only 36% of the users has the location enabled:
67621 false
38308 true

The 10 first location are:
9122 India
3401 New Delhi, India
2293 Mumbai, India
1198 Mumbai
1057 भारत
986 Bengaluru, India
936 New Delhi
636 Pune, India
605 Hyderabad, India
587 india

The most active accounts are:
299 @Amanatlani1
136 @bhattjikamal
125 @chetanzade
100 @Dhananj81918423
97 @pathan_modassar
97 @gurjarss
91 @HYTEIN

Apparently, @Dhananj81918423 has been suspended recently 😄

In the top 6 of the most active accounts, 4 are from the same community: the BJP troll community

Hello bhakts 👋 @Amanatlani1 @bhattjikamal @chetanzade @Dhananj81918423 @gurjarss @BSBHATIOFFICIAL @i_m_rook @Rumtajogi22 @Ganesh_Prajapat @sumitkrsaurav1 @Amn_singh2102 @rangakush @CvimalsutharA @Ravisin31812573 @hemsaini101 @svmanyam @vjmindian @rajas001 @drbabasaheb1

Let's take an example. By doing a reverse image search on the @sumitkrsaurav1's profile picture we can see that this picture is "Cool boy" profile picture used on Facebook

So what is the real impact of this community on the overall tweet volume. At their maximum, they are responsible of more than 10% of the tweet volume

47373 unique Twitter accounts tweeted. This community is composed of 2629 Twitter accounts => 5% of the accounts

5% of the Twitter accounts involved, generated more than 10% of the tweets made on a specific time range. Seems like a good disinformation weapon

A less powerful community was quite active too. Say hi to the @INCIndia IT Cell 👋

Hi @pathan_modassar @ArshadChishti_ @1mdkhan @Miss_Ziddi313 @Husein21_ @KaranDe76737049 @VazeIndian @Priyank80223499 @jainanil1968 @justTripleR @Mushir95770465 @DosalaMosa @ROFLAiyyar_ @IAmShereAalam @jigyasa203

Are you kidding me @TwitterSupport? @pathan_modassar joined Twitter in August 2018 and made 196,000 tweets. This is an insane number...

The Congress IT Cell is less powerful than the BJP one. At their maximum, they are able to generate only 7% of the overall tweet volume

Conclusion: Between 17% and 20% of the tweet volume was not genuine and made by the IT cell of a political party.

This is what disinformation looks like in the 21st century

.@INCIndia and @BJP4India: I hope you are ready for the elections. On my side, I am and I will expose you one by one.