Cryptographically Private Support Vector Machines

Sven Laur; Helger Lipmaa; Taneli Mielikäinen

Paper 2006/198

Cryptographically Private Support Vector Machines

Sven Laur, Helger Lipmaa, and Taneli Mielikäinen

Abstract

We study the problem of private classification using kernel methods. More specifically, we propose private protocols implementing the Kernel Adatron and Kernel Perceptron learning algorithms, give private classification protocols and private polynomial kernel computation protocols. The new protocols return their outputs---either the kernel value, the classifier or the classifications---in encrypted form so that they can be decrypted only by a common agreement by the protocol participants. We also show how to use the encrypted classifications to privately estimate many properties of the data and the classifier. The new SVM classifiers are the first to be proven private according to the standard cryptographic definitions.

Note: Full version

Metadata

Available format(s): PDF
Category: Cryptographic protocols
Publication info: Published elsewhere. The Twelfth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Keywords: Privacy preserving data mining kernel methods
Contact author(s): lipmaa @ ut ee
History: 2006-06-20: received
Short URL: https://ia.cr/2006/198
License: CC BY

BibTeX

@misc{cryptoeprint:2006/198,
      author = {Sven Laur and Helger Lipmaa and Taneli Mielikäinen},
      title = {Cryptographically Private Support Vector Machines},
      howpublished = {Cryptology ePrint Archive, Paper 2006/198},
      year = {2006},
      note = {\url{https://eprint.iacr.org/2006/198}},
      url = {https://eprint.iacr.org/2006/198}
}