Privacy-preserving boosting

被引：26
作者：
Gambs, Sebastien [1 ]
Kegl, Balazs [1 ]
Aimeur, Esma [1 ]
机构：
[1] Univ Montreal, Dept Comp Sci & Operat Res, Montreal, PQ H3C 3J7, Canada
来源：
DATA MINING AND KNOWLEDGE DISCOVERY | 2007年 / 14卷 / 01期
基金：
加拿大自然科学与工程研究理事会;
关键词：
privacy-preserving data mining; boosting; AdaBoost distributed learning; secure multiparty computation;
D O I：
10.1007/s10618-006-0051-9
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
We describe two algorithms, BiBoost ( Bipartite Boosting) and MultBoost ( Multiparty Boosting), that allow two or more participants to construct a boosting classifier without explicitly sharing their data sets. We analyze both the computational and the security aspects of the algorithms. The algorithms inherit the excellent generalization performance of AdaBoost. Experiments indicate that the algorithms are better than AdaBoost executed separately by the participants, and that, independently of the number of participants, they perform close to AdaBoost executed using the entire data set.
引用
收藏
页码：131 / 170
页数：40
相关论文
共 54 条
[51]   A THEORY OF THE LEARNABLE [J].
VALIANT, LG .
COMMUNICATIONS OF THE ACM, 1984, 27 (11) :1134-1142
[52]   State-of-the-art in privacy preserving data mining [J].
Verykios, VS ;
Bertino, E ;
Fovin, IN ;
Provenza, LP ;
Saygin, Y ;
Theodoridis, Y .
SIGMOD RECORD, 2004, 33 (01) :50-57
[53]  
Yao A. C., 1986, 27th Annual Symposium on Foundations of Computer Science (Cat. No.86CH2354-9), P162, DOI 10.1109/SFCS.1986.25
[54]  
Yu H., 2006, P 2006 ACM S APPL CO, P603, DOI [DOI 10.1145/1141277.1141415, 10.1145/1141277.1141415]
123456