two questions about Twibot22

[Jajison]

Hello, I have two questions about twibot22.
（1）I was wondering how you get such a large amount of data through the twitter api (which seems to have a speed limit). And, if I want to use Twibot22, are there privacy concerns?
（2）In the edge.csv，the source entity of relation 'followed' and 'membership' is list, and the target entity is user, this is different from in the paper，are the two entities reversed?

[BunsenFeng]

Hi, thank you for your interest in our work!

For the first question, yes, all data is collected through the Twitter API. Yes, there is a speed limit, so it took us a long time to collect it. As for privacy concerns, we do not publicly release the TwiBot-22 paper and instead ask researchers to apply with their institutional email address (to verify their identity) and clearly state their use case, which is in line with previous datasets such as Cresci-17 and TwiBot-20. I would advise you and other TwiBot-22 users to only use it for research purposes, do not share it with people outside your project, and refrain from reporting on specific users and their info in the paper or other venues.

For the second question, I would refer to @whr000001, who is responsible for data collection and formatting. Hopefully, he would help clarify this. :)

[whr000001]

Hi, thank you for your interest in our work! For the second question, the 'member' relation means a user is a member of a list, the 'follow' relation means a user follows a list. We make a mistake in the paper and you can refer to the 'edge.csv' file.