A Game Theoretic Perspective Toward Practical Privacy Preserving Data Mining

Analysis of privacy-sensitive data in a multi-party environment often assumes that the parties are well-behaved and they abide by the protocols. Parties compute whatever is needed, communicate correctly following the rules, and do not collude with other parties for exposing third party sensitive data. This paper argues that most of these assumptions fall apart in real-life applications of privacypreserving distributed data mining (PPDM). It offers a more realistic formulation of the PPDM problem as a multi-party game where each party tries to maximize its own objectives. The paper uses this game-theoretic framework for doing equilibrium-analyses of existing PPDM algorithms. It then modifies these algorithms using the concept of mechanism design and shows how introduction of penalty forces dishonest rational participants to follow the protocol. It illustrates this using the secure sum computation protocol. Finally, this paper discusses the open questions in this work and future research directions.