six seven

Recall—that is, how many users were successfully deanonymized—was as high as 68 percent. Precision—meaning the rate of guesses that correctly identify the user—was up to 90 percent.

I take that to mean there is a 90% match between anonymized posts and real life profiles for 68% of users and that it’s a minimum confidence level needed for a user to be considered deanonymized.

67% made a match. 90% of matches were right.

No idea how they got that number, though.

Precision: ratio of true positives to total predicted positives.

Recall: ratio of true positives to actual positives