Privacy-Preserving Naive Bayesian Classification Over Vertically Partitioned Data

Authors

Justin Zhan, School of Information Technology & Engineering, University of Ottawa, CanadaFollow
Stan Matwin, 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

Protection of privacy is a critical problem in data mining. Preserving data privacy in distributed data mining is even more challenging. In this paper, we consider the problem of privacy-preserving naive Bayesian classification over vertically partitioned data. The problem is one of important issues in privacypreserving distributed data mining. Our approach is based on homomorphic encryption. The scheme is very efficient in the term of computation and communication cost.

Recommended Citation

Zhan, Justin; Matwin, Stan; and Chang, LiWu, "Privacy-Preserving Naive Bayesian Classification Over Vertically Partitioned Data" (2005). ICEB 2005 Proceedings (Hong Kong, SAR China). 84.
https://aisel.aisnet.org/iceb2005/84