Predicting trending topics on Twitter with 95% accuracy
Eric Barker stashed this in Tech
"At the Interdisciplinary Workshop on Information and Decision in Social Networks at MIT in November, Associate Professor Devavrat Shah and his student, Stanislav Nikolov, will present a new algorithm that can, with 95 percent accuracy, predict which topics will trend an average of an hour and a half before Twitter's algorithm puts them on the list — and sometimes as much as four or five hours before."
At first I thought they had some kind of sorcery.
But really they just have a more efficient tweet analysis algorithm.
Too bad there's no way to use it to make money!
There are a few:
Sell it to Twitter.
Sell the data to traders and analysts.