Title :
Privacy-Preserving Classification on Horizontally Partitioned Data
Author :
Tian, Tian ; Hua, Duan ; Guoping, He
Author_Institution :
Coll. of Inf. Sci. & Eng., Shandong Univ. of Sci. & Technol., Qingdao, China
Abstract :
With the appearance of large-scale database and people´s increasing concern about individual privacy, privacy-preserving data mining becomes a hot study area, to which the support vector machine(SVM) belongs. In this paper, a novel privacy-preserving SVM for horizontally partitioned data is given. It has comparable accuracy to that of an ordinary SVM as we obtain the SVM by using the distinct property of the orthogonal matrices.
Keywords :
data mining; data privacy; database management systems; matrix algebra; pattern classification; support vector machines; horizontally partitioned data; large-scale database; orthogonal matrix; privacy-preserving classification; privacy-preserving data mining; support vector machine; data mining; horizontally partitioned data; orthogonal matrix; privacy-preserving; support vector machine;
Conference_Titel :
Computational Intelligence and Security (CIS), 2010 International Conference on
Conference_Location :
Nanning
Print_ISBN :
978-1-4244-9114-8
Electronic_ISBN :
978-0-7695-4297-3
DOI :
10.1109/CIS.2010.56
