Protection of privacy is one of important problems in data mining. The unwillingness to share their data frequently results in failure of collaborative data mining. This paper studies how to build a decision tree classifier under the following scenario: a database is horizontally partitioned into multiple pieces, with each piece owned by a particular party. All the parties want to build a decision tree classifier based on such a database, but due to the privacy constraints, neither of them wants to disclose their private pieces. We build a privacy-preserving system, including a set of secure protocols, that allows the parties to construct such a classifier. We guarantee that the private data are securely protected.
Zhan, Justin; Matwin, Stan; and Chang, LiWu, "Privacy-Preserving Decision Tree Classification over Horizontally Partitioned Data" (2005). ICEB 2005 Proceedings. 70.