it's important to know that Twitter has two classes of poster: people over 10K followers, and everyone else.
if you're not an influencer, you might as well be just a tweet
kdd.org/kdd2020/accept
Conversation
basically the way this works is, if you're in the top 1% of Twitter users by following count, You're an influencer. The rest of us users are classified into SimClusters based on which influencers we follow in common.
This clustering is done before tweets are delivered to users.
1
1
23
That's what they mean by a heterogeneous representation: Your TL is literally different than other peoples. You're trapped in a bubble based on which influencers you follow.
This creates the phenomenon known as "ingroup", among others
2
1
25
and THAT is why I need 10,000 followers. So I can create a SimCluster of my own, with my cool friends, and we can all hang out together. including you!
pray for me
twitter.com/ELFonthebeatMG
This Tweet is unavailable.
5
28
Replying to
How do you tell if you have your own SimCluster, actually, what can you do if you do vs if you don't? >_>
1
1
or rather, if you're an influencer (what node type is this), since a SimCluster seems to be a cluster that mostly follows a specific *set* of influencers?
1
2
Replying to
influencers are v-nodes, and (by my very back-of-the-envelope calculation) people under 10K followers. is that what you were asking?
1
1
Replying to
You mean over 10K? I guess I was asking how people concretely benefitted from having that many followers?
1
2
a friend's experience tho
Quote Tweet
Replying to @deepfates
Several aspects of the twitter user experience changed once I was over 10k. You can't see the granular follow count, certain types of reply just don't show up, despite a lack of filters. The whole app seems to rearrange itself to isolate you from what it considers to be noise.
1
1
Is the threshold still 10k? Sounds like they might tune it regularly and also progressively complicate it. Like google’s algorithm evolved significantly from of pagerank within a few years.
Basic principles probably still stand though.
it's very much an estimate, based on numbers in the paper and the power law dynamics of Twitter users. and i made that estimate back in 2020 when the paper came out, so i don't know how much those numbers have changed since then. but a useful heuristic i think
1



