Wondering if anyone has an answer to the question:
After finishing this project, consider coming back and playing with this threshold number. How do you think your model would work if it was trying to find incredibly viral tweets? For example, how would it work if it were looking for tweets with 1000 or more retweets?
From the project:
Twitter Classification Cumulative Project: Viral Tweet
Defining Viral Tweets section
I would like to compared my answer to the question to someone else answer.