Abstract: Knowledge extraction from sensitive data often needs collaborative work. Statistical databases are generated from such data and shared among various stakeholders. In this context, the ownership protection of shared data becomes important. Watermarking is emerging to be a very effective tool for imposing ownership rights on various digital data formats. Watermarking of such datasets may bring distortions in the data. Consequently, the extracted knowledge may be inaccurate. These distortions are controlled by the usability constraints, which in turn limit the available bandwidth for watermarking. Large bandwidth ensures robustness; however, it may degrade the quality of the data. Such a situation can be resolved by optimizing the available bandwidth subject to the usability constraints. Optimization techniques, particularly bioinspired techniques, have become a preferred choice for solving such issues during the past few years. In this paper, we investigate the usability of various optimization schemes for identifying the maximum available bandwidth to achieve two objectives: (1) preserving the knowledge stored in the data; (2) maximizing the available bandwidth subject to the usability constraints to achieve maximum robustness. The first objective is achieved with a usability constraint model, which ensures that the knowledge is not compromised as a result of watermark embedding. The second objective is achieved by finding the maximum bandwidth subject to the usability constraints specified in the first objective. The performance of optimization schemes is evaluated using different metrics.

优化算法在所有权保留数据挖掘中的应用

[1]Agrawal R, Kiernan J, 2002. Watermarking relational databases. Proc 28^th Int Conf on Very Large Databases, p.155-166.

[2]Atallah MJ, Raskin V, Hempelmann CF, et al., 2002. Natural language watermarking and tamper-proofing. Int Workshop on Information Hiding, p.196-212.

[3]Bertino E, Ooi BC, Yang Y, et al., 2005. Privacy and ownership preserving of outsourced medical data. Proc 21^st Int Conf on Data Engineering, p.521-532.

[4]Duran MA, Grossmann IE, 1986. An outer-approximation algorithm for a class of mixed-integer nonlinear programs. Math Programm, 36(3):307-339.

[5]Elbeltagi E, Hegazy T, Grierson D, 2005. Comparison among five evolutionary-based optimization algorithms. Adv Eng Inform, 19(1):43-53.

[6]Engelbrecht AP, 2007. Computational Intelligence: an Introduction (2^nd Ed.). Wiley, New York.

[7]Franco-Contreras J, Coatrieux G, 2015. Robust watermarking of relational databases with ontology-guided distortion control. IEEE Trans Inform Forens Secur, 10(9):1939-1952.

[8]Franco-Contreras J, Coatrieux G, Cuppens F, et al., 2014. Robust lossless watermarking of relational databases based on circular histogram modulation. IEEE Trans Inform Forens Secur, 9(3):397-410.

[9]Grossmann IE, Kravanja Z, 1997. Mixed-integer nonlinear programming: a survey of algorithms and applications. In: Biegler LT, Coleman TF, Conn AR, et al. (Eds.), Large-Scale Optimization with Applications. Part II: Optimal Design and Control. Springer, New York, NY, p.73-100.

[10]Hartung F, Kutter M, 1999. Multimedia watermarking techniques. Proc IEEE, 87(7):1079-1107.

[11]Holland, JH, 1992. Adaptation in Natural and Artificial Systems: an Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. MIT Press, Cambridge, MA.

[12]Iqbal S, Rauf A, Mahfooz S, et al., 2012. Self-constructing fragile watermark algorithm for relational database integrity proof. World Appl Sci J, 19(9):1273-1277.

[13]Kamran M, Farooq M, 2012. An information-preserving watermarking scheme for right protection of EMR systems. IEEE Trans Knowl Data Eng, 24(11):1950-1962.

[14]Kamran M, Farooq M, 2013. A formal usability constraints model for watermarking of outsourced datasets. IEEE Trans Inform Forens Secur, 8(6):1061-1072.

[15]Kamran M, Suhail S, Farooq M, 2013. A robust, distortion minimizing technique for watermarking relational databases using once-for-all usability constraints. IEEE Trans Knowl Data Eng, 25(12):2694-2707.

[16]Kennedy J, Eberhart R, 1995. Particle swarm optimization. Proc IEEE Int Conf on Neural Networks, p.1942-1948.

[17]Khanduja V, Verma OP, 2012. Identification and proof of ownership by watermarking relational databases. Int J Inform Electron Eng, 2(2):274-277.

[18]Khanduja V, Chakraverty S, Verma OP, 2015. Watermarking categorical data: algorithm and robustness analysis. Def Sci J, 65(3):226-232.

[19]Rani S, Kachhap P, Haider R, 2016. Dataflow analysis-based approach of database watermarking. In: Chaki R, Cortesi A, Saeed K (Eds.), Advanced Computing and Systems for Security, Volume 2. Springer, New Delhi, p.153-171.

[20]Rao UP, Patel DR, Vikani PM, 2012. Relational database watermarking for ownership protection. Proc Technol, 6:988-995.

[21]Schrage LE, 1991. LINDO: an Optimization Modeling System (4^th Ed.). Scientific Press, South San Francisco, CA.

[22]Shehab M, Bertino E, Ghafoor A, 2008. Watermarking relational databases using optimization-based techniques. IEEE Trans Knowl Data Eng, 20(1):116-129.

[23]Sion R, Atallah M, Prabhakar S, 2004. Rights protection for relational data. IEEE Trans Knowl Data Eng, 16(12): 1509-1525.

[24]Wang YM, Gao YX, 2012. The digital watermarking algorithm of the relational database based on the effective bits of numerical field. World Automation Congress, p.1-4.

[25]Wolfgang RB, Delp EJ, 1996. A watermark for digital images. Proc 3^rd IEEE Int Conf on Image Processing, p.219-222.

[26]Wylie JE, Mineau GP, 2003. Biomedical databases: protecting privacy and promoting research. Trends Biotechnol, 21(3):113-116.

[27]Zhang LZ, Gao W, Jiang N, et al., 2011. Relational databases watermarking for textual and numerical data. Int Conf on Mechatronic Science, Electric Engineering and Computer, p.1633-1636.