Infiltrating the #filterbubble?
Slandering
Donald Trump and telling racist jokes seems to be the intention of a social
botnet on Twitter.
There have
been different stories about bot armies supporting Donald Trump.
@PatrickRuffini has posted a list with bots that are likely supporting Trump.
@Wired has wrote about pseudo Hispanic bots, supporting Trump. But Donald Trump is also victim of
slandering bots. Morteza Shahrezaye (@morry_shah) and I have found a botnet with a strange
behavior: The bots are telling jokes all the time of which many are very
racist, disturbing anti-Semitic and sexist. In addition, the bots post “jokes”
about Donald Trump, like: “Your moma is so dump, she voted for Trump.”
There are
at least three possible explanations for this:
- There might be a racist botmaster who thinks Trump is too liberal.
- It could be a “fluke”: The bots are just there to tell jokes but the internet is full with racist stuff and anti-Trump jokes, so this is just what the bots find.
- Someone is following a more complicated strategy.
We think
number 3 is right. In our opinion, this strange behavior makes a lot of sense.
If the botmaster is assuming Trump fans to be in a Trump-filterbubble,
slandering Trump on Twitter won’t have any effect. The users won’t get in touch
with the Trump-jokes and if they do by chance, they will see directly that it
is part of an anti-Trump campaign. But if you could get inside the
filterbubble, things might change. Racist jokes might be a way to get a
positive feedback from Trump-supporters. Note: I am not saying,
Trump-supporters are racists. I am only saying that someone who wants to work
against Trump might think that his supporters are racists.
Of course,
we have some data that supports this theory:
Bot-Birthdays |
Here you
see the “birthday” of the bots. They are all very young which might support the
idea that the whole thing has to do something with the elections.
Density of Retweets |
Looking at
the density of the average retweet count for tweets about Trump (blue) and all
other jokes you see that it follows a different pattern. Some of the Trump
jokes are retweeted more often than all other jokes (we have a heavy tail
distribution). Look at the density of "Jew-jokes" (green), as well.
Wordcloud (unique messages) |
Wordcloud (all messages) |
Looking at
a wordcloud of the most frequent words you can find “Trump” as well within the
unique messages (the bots are repeating the same jokes over and over again) as
within the data for all the tweets (including duplicates). But we do not find
any other names of politicians or celebrities. Ergo: Slandering Trump is
probably the real intention of this botnet.
But be
aware: In time of big data very strange things might occur frequently by
chance!
And for all the data-scientists here comes some sample data on the bots:
BOTS.screen.Names BOTS.N BOTS.N.trump BOTS.N.JEW BOTS.friends BOTS.follower averageRTcount averageRTcountTRUMP averageRTcountJEW
a4hkvlasov a4hkvlasov 2945 47 54 218 16 0.01052632 0.0212766 0.01851852
AmyBirdiesl AmyBirdiesl 354 6 11 18 0 0.01977401 0 0
anatoliyg90zi anatoliyg90zi 1050 12 19 63 12 0.01333333 0 0
artemd6ye artemd6ye 3200 54 47 404 69 0.0096875 0 0.02083333
ChloeWilhelmina ChloeWilhelmina 349 5 8 19 2 0.00286533 0 0
deniscdcershov1 deniscdcershov1 3200 44 31 402 31 0.005 0.02272727 0.03225806
eduardzuev68o eduardzuev68o 3200 51 53 371 47 0.0140625 0.01960784 0
EllaTilliema EllaTilliema 427 8 11 31 4 0.009367681 0 0
EmilyJaquelineo EmilyJaquelineo 504 7 12 35 6 0.01984127 0 0
evgurev7r evgurev7r 2817 39 54 176 53 0.01561945 0.02564103 0.03703704
IlyasermiAshley IlyasermiAshley 3200 57 49 355 38 0.0096875 0.03508772 0.02040816
ivanozrgalkin1 ivanozrgalkin1 2927 54 42 209 23 0.0102494 0.01851852 0
JasmineCariha JasmineCariha 429 5 11 28 9 0.006993007 0 0
JessicaYongoe JessicaYongoe 287 6 4 19 3 0.01045296 0 0
KanishaChasity KanishaChasity 3200 52 44 378 58 0.00625 0.03846154 0
kirilltevor kirilltevor 1006 8 20 57 7 0.01590457 0 0.05
konstantinfk9k konstantinfk9k 1004 5 21 65 17 0.01992032 0 0.04545455
KristyIndia KristyIndia 3117 58 50 247 24 0.01090792 0 0.03921569
MacyZetta MacyZetta 3200 45 46 413 73 0.010625 0.04444444 0.02173913
monica_shanice monica_shanice 3200 54 45 363 35 0.011875 0.03703704 0
oleg0ywns oleg0ywns 619 12 15 47 3 0.008077544 0 0
OliviaLatricema OliviaLatricema 291 3 4 30 8 0.04467354 0 0.25
pavel1s1ckarpo1 pavel1s1ckarpo1 623 7 6 44 7 0.01284109 0 0
romanbe7lim romanbe7lim 621 13 11 51 2 0.01610306 0 0
sashasharovk9k sashasharovk9k 3200 55 39 349 42 0.010625 0.05454545 0
sashaurax sashaurax 2858 42 51 230 27 0.01119664 0 0
sergeywoov sergeywoov 2808 38 36 212 39 0.01103989 0 0
stepanpanovjk stepanpanovjk 3200 48 39 374 51 0.0096875 0.02083333 0
TaniaGiusy TaniaGiusy 3200 60 48 390 51 0.01625 0.01666667 0.02083333
valentin9tk valentin9tk 624 14 9 47 4 0.01442308 0 0
vasiliykonqigh vasiliykonqigh 3199 34 47 336 21 0.005939356 0.02941176 0.0212766
vitaliysok1p vitaliysok1p 3112 45 57 238 31 0.007390746 0 0.03508772
vyacheslavgajhb vyacheslavgajhb 2878 45 55 219 42 0.0114663 0.02222222 0
yaroslav3pwnme1 yaroslav3pwnme1 3200 40 44 394 35 0.01 0.05 0
tarynnuria TarynNuria 3200 55 33 223 30 0.00875 0 0
kristyindia KristyIndia 2673 47 43 216 21 0.01084923 0 0.04545455
ChloeArleenes ChloeArleenes 720 9 18 58 8 0.009722222 0 0.05555556
corinne_rosa corinne_rosa 3129 63 55 232 37 0.01278364 0.03174603 0
EllaRosaleel EllaRosaleel 486 10 9 41 3 0.008230453 0 0
JessicaDestin1 JessicaDestin1 726 7 13 67 8 0.01928375 0 0
patriciadschil4 patriciadschil4 465 7 8 35 2 0.0172043 0.1428571 0
valentingusp01 valentingusp01 3199 56 43 436 59 0.006564551 0.05357143 0.02325581
joy_hyo joy_hyo 3200 53 45 430 59 0.035625 0.05660377 0.02173913
lAmFunnyBot lAmFunnyBot 3195 52 54 23 23 0.009389671 0.09615385 0
HannahRethala HannahRethala 722 12 12 56 8 0.02077562 0.08333333 0.08333333
sandrasstoner1 sandrasstoner1 442 8 5 29 3 0.01357466 0 0
not_actuallyme not_actuallyme 3200 26 28 1 166 0.0028125 0 0
yuriykox0a yuriykox0a 279 6 2 37 2 0.007168459 0 0
IStochliya IStochliya 411 5 4 54 10 0.00973236 0 0
williejstackt7 williejstackt7 449 10 4 34 0 0.01113586 0 0
SarahTrevaom SarahTrevaom 564 16 7 30 6 0.008865248 0 0
CharlotteMauri CharlotteMauri 492 6 6 35 5 0.01219512 0 0
BestoReddit BestoReddit 3200 16 12 0 33 0.0046875 0.0625 0.08333333
BeverlyNicholes BeverlyNicholes 1706 14 18 120 18 0.005861665 0 0.05555556
GltrogovThomas GltrogovThomas 1333 18 20 76 6 0.01350338 0 0
richarddearll richarddearll 1120 20 16 71 13 0.01160714 0 0
Kommentare
Kommentar veröffentlichen