News for hackers. Stuff that matters.
Re:Quantifying the impact of the Twitter fake accounts purge... by TangoTrotFox on Wednesday August 08, 2018 @12:14PM (#17717136)
I have to say I rather enjoyed the snide, "Determining how those accounts were able to make such an impressive recovery is left as an exercise to the reader." in relation to users that saw large number of bot followers previously banned, followed by them rapidly gaining huge numbers of new followers... somehow.
Re:Quantifying the impact of the Twitter fake accounts purge... by AndrewStephens on Wednesday August 08, 2018 @01:33PM (#17717741)
I don't imagine the creators of the bot accounts were too upset by Twitter eventually banning millions of their creations. Now they get to sell millions of additional accounts to the same people all over again.

It doesn't take any fancy machine learning to figure out what is happening.

Re:Quantifying the impact of the Twitter fake accounts purge... by jewelthief91 on Wednesday August 08, 2018 @06:51PM (#17720308)
Brilliant. Twitter gets let off the hook by appearing to be "doing something" and people who charge for followers get a revenue boost by this new form of planned obsolescence. Twitter users get screwed by either getting caught up in the purges or having to pay even more money to bot creators to inflate their numbers, likely pricing many out of the market. Seems like a good time to short Twitter stock...
Re:Quantifying the impact of the Twitter fake accounts purge... by alwayseasy on Wednesday August 08, 2018 @04:22PM (#17719125)
My guess: many humans were caught in the purge. I've seen people explain that Twitter locked their account, asking for a phone number for verification. While they were locked out, they were subtracted from the follow count of the accounts they followed.
Re:Quantifying the impact of the Twitter fake accounts purge... by btilly on Wednesday August 08, 2018 @05:27PM (#17719654)
My guess, new fake accounts. The best evidence for that is the following snippet from the article:

The most striking example of those massive rebounds is the account of @AdelAliBinAli. Before the purge, this account had 6.2M followers. On July 14, its follower count was down to 2.7M followers (-3.4M / -55%). Between July 14 and July 19, it regained 1.2M followers, and as of July 23rd, this account now has 6.6M followers, which is even more than before the purge and represents 3.9M additional followers gained.

As for the others, having more than a statistically average number of purged followers is evidence that you were buying fake followers in the first place. The obvious explanation for large increases in followers for someone who is already known to buy them is that they bought more.