Title :
Classification of Privacy-preserving Distributed Data Mining protocols
Author :
Xu, Zhuojia ; Yi, Xun
Author_Institution :
Sch. of Comput. Sci. & Math., Victoria Univ., Melbourne, VIC, Australia
Abstract :
Recently, a new research area, named Privacy-preserving Distributed Data Mining (PPDDM) has emerged. It aims at solving the following problem: a number of participants want to jointly conduct a data mining task based on the private data sets held by each of the participants. This problem setting has captured attention and interests of researchers, practitioners and developers from the communities of both data mining and information security. They have made great progress in designing and developing solutions to address this scenario. However, researchers and practitioners are now faced with a challenge on how to devise a standard on synthesizing and evaluating various PPDDM protocols, because they have been confused by the excessive number of techniques developed so far. In this paper, we put forward a framework to synthesize and characterize existing PPDDM protocols so as to provide a standard and systematic approach of understanding PPDDM-related problems, analyzing PPDDM requirements and designing effective and efficient PPDDM protocols.
Keywords :
data mining; data privacy; protocols; PPDDM requirements; information security; privacy preserving distributed data mining protocols; private data sets; Classification algorithms; Cryptography; Data models; Data privacy; Distributed databases; Protocols; Privacy; classification; distributed data mining;
Conference_Titel :
Digital Information Management (ICDIM), 2011 Sixth International Conference on
Conference_Location :
Melbourn, QLD
Print_ISBN :
978-1-4577-1538-9
DOI :
10.1109/ICDIM.2011.6093356