Investigative team Sadbottrue has own technologies for data gathering and for meta-analysis of big data. We focused on findings and explanation of anomalies that influence on the tech companies. We are fighting against the social media and advertising fraud ecosystem.
Chapter 31. Brexit VS StrongerIn
The true story
A motorcyclist, a Dali clock, a doghead woman, a gym, a cherry, a bee, a bronze Churchill, two cats, an airplane, a flower, a frog, a sheep, Popeye sailor, another one bronze Churchill, a lighthouse, a shepherd, a dirigible, berries, a black square, Dalai Lama, a penguin. It is not a meaningless set of words to memory training, as many might think.There are 21 active #Brexit supporters, defending its position by retweeting@vote_leave.
@mariedd52another one bronze Churchill
As in the case of candidates for the Mayor of London, we have identified the most active retweeters in each account and analyzed them. The first part of #Brexit of supporters you have seen, they have nothing in common with real people. This can be said even without a deep analysis of its activity, just studying the clearance and description of the accounts. In fact, after drawing up a set of these accounts, there was a question, and whether to do further analysis. There is extensive use of bots retweeter, and not very disguised, to simulate the high activity of the audience. But despite receiving a final conclusion at the beginning of the study, we decided to continue. We can meet even more interesting facts that are already usual for us, but which will be interested to know our readers.Analysis bots activity can lead us to quite amusing facts. For example, account@UKmehedyis undecided, it retweeted both@vote_leaveand@StrongerIn.Interests of some accounts are not limited to the UK. You can also find here the US Republicans.@crazycatz2013is pro-Trump retweeter.
Here, one might wonder. How the dark side of the ecosystem should be arranged so might appear such intersection?
Most bots are simple retweeters with no other activity. However, there are exceptions. For example, a bot @Fight4UK, who posted a great number of photos with the hashtag. It is also a kind of statistical fraud affecting the trending. @BranVeroPRis the most faithful bot. It is not interested in anything except the referendum, and it clearly defined its position. 45% of its tweets are retweets of@vote_leave.We already wrote anarticleabout bots with duplicate stickers, which was quoted by Trump. Among the #Brexit supporters, we found the same@Julie34479. Can you imagine a woman who twice has stuck herself logos directly on the face one over the other?
Finally, we show you something that will blow your mind. We would never have thought that we could find here an account with photos of a naked old man @steffanrhys3. We especially posted an inactive link, so you do not accidentally click. You'd better not to see it, trust us. Creating a fake Twitter account is low paid and monotonous work. To somehow cheer up themselves, employees create silly and ridiculous in their view profiles, because all of them equally will not be seen by anyone. And sometimes they get lazy, and they even don’t edit the accounts. This way the egg bots appears. This list contains 12 such accounts.
All these accounts increase the number of retweets not only the main @vote_leave but hundreds of other accounts of its group. Many of them are official and unofficial accounts of the Conservative Party and its members, calling to vote for the exit from the EU. Just 3 accounts of top 100@vote_leaveretweeters are exactly humans, and all of them are conservative politicians. Is this a coincidence? We don’t think so. In this list, only verified accounts are exactly the people.
@DCBMEPConservative MEP for Eastern England. Co-chair, Conservatives for Britain. Pro Brexit. Author: 'Time to Jump'. (The presence of this person in this list does not cause questions).
@DouglasCarswellClacton. Running. Gardening. Free trade. Optimist. @Vote_Leave team. Ukip parliamentary party.
The average number of tweets per account from the Top 100 list is 48,000.
93 accounts of top 100@vote_leaveretweeters are exactly bots. Sad, but true.3 accounts are verified, another 4 accounts were classified as cyborgs, that means bots, partly with human activity. Even if we consider them as real people, the automatic activity among #Brexit supporters is 93%. We see the same in the groups with the number of retweets less than 100.
Has the active audience of the account @StrongerIn other distinctive features, or we will see again the same thing? Let's start with the basic statistics. There are 114 most active accounts (more than 100 retweets per account).Thetableof all the accounts with the notes and behavioral classification. Those supporters can be divided into few groups according to their behavior patterns, sub-interests and groups they try to imitate.The main behavior pattern is “bot-retweeter”, that use retweets of “IN” official accounts and “IN” media publication for its own twitter feed. But among this group there are sub-groups.
The first one is egg-bots. Ordinary, primitive bots for enlarging of the “IN” official accounts retweet count. A couple of eggs try to be trolls, but even Humpty Dumpty is better in a chitchat.
The second group is bot-retweeters with animal profile. Creators of the botnet responsible attitude to the job, and chose the majestic animals.
Human’s account with bot-retweeter’s behavior pattern is the largest part of the botnet. Bots with the fake identity, stolen photos or automatised accounts of real people.
The first subcategory is a classic fake supporter. The profile has the photo of the real human, targeted sources of retweets, but it has no other interests except “IN” campaign. Targeted fake supporters is a common manipulation technique. For example @FickersH targeted for Christian conservators.
The most interesting is the account @jgunson1989. It is serial retweeter, who has switched from one source to another after a set of retweets. Sometimes it uses different sources united by one news.
The account @falcos2012 focused on pop-science news. Other accounts use targeted news about Brexit for filling it’s own feed. This accounts are created by the real people for business or personal promotion and are involved in the “IN” campaign.
@PimlicoPlumbers the account is the hybrid between “IN” campaign and plumper service company. @_laurenpn_ Intersectional feminist and Youth Representative for Ethnic Minority with a lot of retweets of beauty and fun content. @samflatman the account of the educational business. @charliemansell is an account of political researcher of www.thecampaigncompany.co.uk
The next big category is the official accounts.
The accounts in this group have proclaimed, that they are an official part of “IN” campaign (both corporate and personal). Those accounts belong to real people. Most of them have a unique content, comments, photos, but the main pattern is “bot-retweeters” (with tens thousands of tweets). All they have bots in its friends. We guess, that local official and semi-official accounts are used in the “IN” botnet.
The last such account is @StrongerInPress. But in fact, it is a most ordinary bot, hiding under the guise of the official account. If you change the picture and description, no one would ever guess that this account was media aggregator.
The next category is cyborg’s accounts. It is complicated behavioral pattern, that has to combine both bot-generated retweets and human-generated content and conversations. Cyborgs are useful for trolling.
Surprisingly, there are also accounts with some sex-related content. If your brain is not exploded in the last time, it is sure to explode right now. All links can be clicked. Most accounts of sex-and-fun-addicted college students, gay-socialists, and even nudist-monarchist. Not safe for viewing at work.
Summarizing, we can say that among Top 100 most active @strongerin retweeters there are just 12 accounts belong to human. 88 are bots, 26 among them have the signs of humans (photos, unique comments). The average amount of tweets per the account is 28,300.
Is this situation is something extraordinary for social media campaigns? More likely “no” than “yes”. This is the stereotype that all major events should have social coverage, and always with a high level of engagement. Social media experts and community managers are given the task to get a certain result in the social networks. For this they are accountable, and that is what they are paid.
It is a misconception that the absence of a huge number of likes/retweets/shares on social networks means that people don’t care about that. In fact, really caring people will discuss the topic, and to bring their arguments and not just formal retweet/repost slogans. And the activity of the audience of the official account is only required for the reporting the head of the Social Media campaign. But there is another reason why it is done.
Imagine that one side has the real number of organic retweets, for example, 7% of the current amount. And this, incidentally, would be a good result. But the opposite party has used the possibility of automatic cheating, and showed 10-15 time higher engagement than the opponents. What would the next day all the journalists have written? We are confident that all the newspapers would write that one party completely losing campaign in social networks, and it would certainly have the impact the results of the voting.
As a result, both sides need to play by the rules established the social networks. They enjoy the unlimited confidence, and their audience is without question recognized as real people. Somehow, no one can check whether this is so. But check the reliability of the facts, it is the primary duty of journalists. Perhaps it is time to show at least a little critical thinking.
The bot facts
1. It might be said that the most active audience @StrongerIN looks a little better if the general picture in both accounts would not be so terrible.
2. Among 200 most active “Brexit / StrongerIn” retweeters there are only 19 human’s accounts.
3. Both “Brexit / StrongerIn” campaign managers use the same tricks with the fake audience. In fact, the real influence is very weak, because of the huge amount of the fake followers among the common audience.
The sad question
Why none of the social media analytics company did the report on most active retweeters at least one of a political campaign, and to verify these accounts on bot activity?