pump-and-dump-dataset
pump-and-dump-dataset copied to clipboard
Question: How was pump_telegram.csv populated?
Hello,
I find your work fascinating and I'm incredibly appreciative of you making not just the paper available but also much of the data, so thank you for that.
I did wonder though; how were you able to populate the raw telegram announcement data?
Was it through simple pattern matching, manual work or maybe some fancy machine learning? If there was any programmatical method used I'd love to see how you overcame the disparity in the "buy X coin now" inbound messages from the pump channels.
Thanks J