臺灣博碩士論文加值系統

在現今的社會中,有釵h隱密性的資料是需要被保護的;然後,有某部分的資料卻是需要被分享合作來達到更大的效益.舉例來說：銀行系統中的顧客交易資料,基因序列的長度比較,疾病診斷的分類.為了達到這種計隱密又公開的性質,安全計算就是為了符合這個特性而提出.
在我們的論文中,我們是用水平的資料分割來完成隱密資料的分類.我們的分類模型適用桿知機類神經網路模型.對於不同的啟動函式,我們根據之前的文獻提出相對應的方法.在我們的論文中,我們提供了一個安全的學習演算法來達成我們的目的.

Today, there are lots of confidential information needing to be protected. However, some of them are required to share for collaboration. For example, query of database in bank account system, sequences comparison for genetic sequence, and classification for diagnosis analysis, etc. In order to achieve confidentiality and collaboration, secure computation had been proposed from years ago.
In our thesis, we try to achieve classification of confidential data which are horizontal partition. Our model for classification is perceptron network, and we propose several methods for related active function. Based on experiences of early papers and updating rules in perceptron. We have secure learning algorithm to accomplish our goal

1 Introduction 4
1.1 Secure Computation . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Perceptron Network . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Thesis Organization . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Preliminaries 8
2.1 Secure Two/Multiple Party Computation . . . . . . . . . . . . 8
2.2 Models of Secure ComputationMultiplication . . . . . . . . . 9
2.2.1 Secure Two-Party Scalar Product Protocol 1 . . . . . . 10
2.2.2 Secure Two-Party Scalar Product Protocol 2 . . . . . . 10
2.3 Secure Computation of Sigmoid Function . . . . . . . . . . . . 12
2.4 Extension to SecureMulti-party Computation . . . . . . . . . 14
3 Secure Learning Algorithm 18
3.1 Models of Neural Networks . . . . . . . . . . . . . . . . . . . . 18
3.1.1 Active Function . . . . . . . . . . . . . . . . . . . . . . 18
3.1.2 Learning Rule . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 Single Layer Perceptron with Linear Function . . . . . . . . . 22
3.3 Single Layer Perceptron with Sigmoid Function . . . . . . . . 25
3.4 Single Layer Perceptron with Hard Limit Function . . . . . . . 28
3.5 Multiple Layer Perceptron . . . . . . . . . . . . . . . . . . . . 30
4 Experiment 33
4.1 Data Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.2 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . 34
4.2.1 Single layer Perceptron . . . . . . . . . . . . . . . . . . 34
4.2.2 Multiple layer Perceptron . . . . . . . . . . . . . . . . 36
5 Conclusions and Future Work 41
5.1 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5.2 FutureWork . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
A . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
B . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
C . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
D . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
E . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
F Some Inference in Multiple Layer Perceptron . . . . . . 58

[1] B. Barak, R. Canetti, Y. Lindell, R. Pass and T. Rabin. Secure Computation
Without Authentication. In CRYPTO 2005, Springer-Verlag (LNCS
3621), pages 361-377, 2005.
[2] C.Cachin. Efficient private bidding and auction with an oblivious third
party. In Proceedings of the 6th ACM conference on Computer and communications
security, November 1999
[3] C. L. Lawson, R. J. Hanson. Solving Least Squares Problems. Englewood
Cliffs, N.J.: PrenticeHall, pp. 180-198, 1974.
[4] D.Chaum, C.Crepeau, and I.Damgard. Multiparty unconditionally secure
protocols. In Proceedings of the 20th Annual ACM Symposium on Theory
of Computing, STOC’88, 1988.
[5] Goldreich, O. Foundations of Cryptography, Volume II: Basic Applications.
Cambridge University Press, 2004.
[6] M. Atallah, J. Rice. Secure Outsourcing of Scientific Computations. In
Privacy Enhancing Technologies : 63-78, 2004.
[7] Martin T.Hagan, Howard B. Demuth, and Mark Beale. [Book] Neural
Network Desing.
[8] Mikhail J. Atallah, Florian Kerschbaum, Wenliang Du. Secure and Private
Sequence Comparisons. In Proceedings of the ACM workshop on
Privacy in the electronic society, Pages: 39 - 44, 2003
[9] Mikhail J. Atallah, Wenliang Du. Secure Multi-party Computational Geometry.
In roceedings of the 7th International Workshop on Algorithms
and Data Structures, Pages: 165 - 179.
[10] M.Hernandez and S.Stolfo. The merge/purge problem for large
databases. n Proceedings of the 1995 ACM SIGMOD, May 1995.
[11] Pradeep Ravikumar, William W. Cohen. A Secure Protocol for Computing
String Distance Metrics. n Proceedings of ICDM Workshop on
Privacy and Security Aspects of Data Mining, 2004.
[12] Ronald Cramer, and Ivan Damgard. Multiparty Computation, an Introduction.
[13] S. Goldwasser and Y. Lindell. Secure Computation Without Agreement.
In the Journal of Cryptology, 18(3):247-287, 2005.
[14] Wenliang Du .A Study of Several Specific Secure Two-Party Computation
Problems. PhD thesis, Purdue University, West Lafayette, Indiana,
2001.
[15] Yao, A. C. Protocols for Secure Computation. In 23rd IEEE Symposium
on Foundations of Computer Science, pages 160–164, 1982.
[16] Yingpeng Sang. Hong Shen. Zonghua Zhang. An Efficient Protocol for
the Problem of Secure Two-party Vector Dominance. Sixth Int’l Conf.
on Parallel and Distributed Computing, Applications and Technologies
(PDCAT 2005), 488-492, 2005
[17] Y. Ishai, E. Kushilevitz, Y. Lindell and E. Petrank. Black-Box Constructions
for Secure Computation. In the 38th STOC, pages 99-108, 2006.
[18] Y. Lindell. Parallel Coin-Tossing and Constant-Round Secure Two-
Party Computation. In the Journal of Cryptology, 16(3):143-184, 2003.
[19] http://www.pinkas.net/ot.html