Privacy-preserving data mining via secure multiparty computation

• Main Author: Han, Shu Guo
• Other Authors: Ng Wee Keong
• Format: Theses and Dissertations
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Computer science and engineering::Information systems::Database management
• Online Access: https://hdl.handle.net/10356/41834
• Institution: Nanyang Technological University

Conventional data mining algorithms handle with the data sets that are usually maintained in one central server. If data sets are distributed among multiple parties, one trusted server collects the data sets first before performing data mining tasks. Distributed data mining (DDM) [27, 28, 103] was proposed to mine the distributed data without data collection. However, data sets held by each party are allowed to fully access by other parties. In recent advances, privacy issues become more and more important, especially when the data are involved with sensitive information. New algorithms are required to mine the data sets distributed among parties while preserving the privacy of each party. Privacy-preserving data mining (PPDM) [9, 74] was proposed to address this issue. Privacy-preserving data mining (PPDM) seeks to utilize distributed data sets that are privately held by individual parties for knowledge discovery while preserving the privacy of their data.