Privacy Preserving Learning of Decision Trees. Benny Pinkas HP Labs Joint work with Yehuda Lindell (done while at the Weizmann Institute). Cryptographic methods. perturbation methods. Cryptographic methods vs. perturbation methods. overhead. This work…. inaccuracy. lack of privacy.
Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.
Privacy Preserving Learning of Decision Trees
Joint work with Yehuda Lindell
(done while at the Weizmann Institute)
a lot of fraud lately…
I can’t find a pattern
to recognize fraud in advance..
Neither can I..
Maybe we should share information..
Have you heard of “Secure
function evaluation” ?
This is all “theory”.
It can’t be efficient.
Wish to “mine”D1D2 without revealing more info
C(x,y) and nothing else
One Exp per log “OT”s [NP]
Secure Function Evaluation [Yao ‘86]
ID3: Choose attribute A that
minimizes the conditional
entropy of the attribute class
> 10 years
Age > 30
Claim > $500
x = e.g. # of patients with (age > 30) and (fraud = yes)
Q( . )
Q(x) and nothing else
Two passes, O(degree) (or O( log|F|) ) exponentiations.
Use Taylor approximation for lnx
ln 2 n + i=1..k(-1) i-1 i / i
= ln 2 n + T()
x =x1 +x2 = 2 n + 2 n
Step 2 of the protocol