Every social media user has the problem of fake followers. Fake followers are annoying. Why? Because they don't engage with your content and that is really bad.
For example, what is the point of having 1000 followers but only 10 of them are active and the rest are fake/inactive followers? It is better if you only have 100 followers but 75 of them are active. You get more for less. Steemit also has this fake follower problem. I see a lot of people following more than 1000 people and I don't believe that they are engaging with the content of their followed people. This is a really bad behavior because Steemit makes a lot more fun if interact with your community.
A few weeks ago, I took a look into the Steemit Python API of @furion. I played with it and then I decided to make a tool to extract your fake followers to get your real followers.
How do I define fake followers?
My first attempt was getting the following count of my followers and look whether this user is following more than 1000 people. But I recognized that there are nice people like @twinner which are following more than 1000 people and interact with the content of the followed people. So I decided to get all my followers; look if they've ever upvoted a post of me and also look if they are following more than 1000 people.
Then I print out:
How many followers never upvoted my content.
How many 1000+ following followers I have and how many of them never interacted with my posts.
Currently, I only look if they upvoted at least one of my posts. I also wanted to see how many followers never wrote a comment but the method of the Steemit API isn't currently working. Another method would be to check only the upvotes of the last x posts and then compare them to my followers to get inactive/fake followers. (Maybe they also didn't like your x last posts.)
This is the whole code. You should be able to run it in a Python environment with the Steem module installed.
If you run the code, you can set a username and then the script does the above described steps. Afterwards it prints the percentage of fake followers.
The results are shocking!
If everything works properly (it should if I'm not dumb), then the results are horrible.
For example, I tested it with my username. It printed the following:
I can't really believe it that so many followers never upvoted a single post of me.
We should vote as much as possible because it's free and we give value to somebody else.
I don't want to witch-hunt anyone so you can't see the fake followers. You can modify the script to enable it. It's pretty easy if you really want to know each fake follower.
What do you think? Does my script properly represent fake followers? Can we classify a fake follower with my ideas? Let's start a discussion in the comments!
If you also want to know your percentage of fake followers, just tell it to me and I'll run the script for you.
P.S: Unfortunately, nobody participated in my Steemit Background Cover Contest :D I will redo the competition at a later time if I have more followers and more influence.
Leave a comment, upvote and a follow if you liked my post. See you soon :)
