How large does can a social circle be before first names no longer suffice for identification? Scott, I'm looking at you.
-
-
Replying to @MakerOfDecision
Considering the top 10,000 personal names, probability of collision = 50% when n<=118 (`ceiling(sqrt(2*10000*log(2)))`)
2 replies 0 retweets 3 likes -
Replying to @gwern
Even worse than that; subgroup name popularity (geographic, socioeconomic) is going to be more concentrated than general population.
1 reply 0 retweets 0 likes -
Replying to @MakerOfDecision
It's an upper bound, yes. Could do better by getting some name datasets & resampling.
1 reply 0 retweets 1 like -
Replying to @gwern @MakerOfDecision
I get 51 as the best estimate: http://www.gwern.net/Statistical%20notes#great-scott-personal-name-collisions-and-the-birthday-paradox …
1 reply 0 retweets 2 likes
MakerOfDecision Retweeted MakerOfDecision
Well, @Gwern has the definitive answer; http://www.gwern.net/Statistical%20notes#great-scott-personal-name-collisions-and-the-birthday-paradox …
It's 118. Or 15. Or, best estimate, 51.https://twitter.com/MakerOfDecision/status/759086911084490752 …
MakerOfDecision added,
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.