Privacy-Preserving Collaborative Association Rule Mining

Authors

Justin Zhan, School of Information Technology & Engineering, University of Ottawa, CanadaFollow
Stan Matwin, School of Information Technology & Engineering, University of Ottawa, CanadaFollow
Nathalie Japkowicz, School of Information Technology & Engineering, University of Ottawa, CanadaFollow
LiWu Chang, Center For High Assurance Computer Systems, Naval Research Laboratory, USAFollow

Document Type

Article

Abstract

In recent times, the development of privacy technologies has promoted the speed of research on privacy-preserving collaborative data mining. People borrowed the ideas of secure multi-party computation and developed secure multi-party protocols to deal with privacy-preserving collaborative data mining problems. Random perturbation was also identified to be an efficient estimation technique to solve the problems. Both secure multi-party protocol and random perturbation technique have their advantages and shortcomings. In this paper, we develop a new approach that combines existing techniques in such a way that the new approach gains the advantages from both of them.

Recommended Citation

Zhan, Justin; Matwin, Stan; Japkowicz, Nathalie; and Chang, LiWu, "Privacy-Preserving Collaborative Association Rule Mining" (2004). ICEB 2004 Proceedings (Beijing, China). 198.
https://aisel.aisnet.org/iceb2004/198