Outsourced Privacy-Preserving Reduced SVM Among Multiple Institutions

Executing data mining algorithms locally is usually computationally intensive. A promising solution is to outsource the heavy data mining tasks and datasets. On the other hand, combining data from multiple institutions for a big and varied training set helps enhance the performance of data mining. Due to privacy concerns, different institutions should encrypt their datasets with different keys. Support Vector Machine (SVM) is a popular classifier. It is challenging to train SVM on encrypted datasets in the cloud. Existing schemes use either the multikey fully homomorphic encryption on one server, or partially homomorphic encryption on two non-colluding servers. The former is inefficient and the institutions have to remain online, while the latter relies too heavily on the assumption of two non-colluding servers. To remove these limitations, we demonstrate how to train SVM for both horizontally and vertically partitioned datasets. To reduce training complexity and enhance security, we focus on reduced SVM with a secure kernel matrix. We proved the security of our scheme and the experimental results validated its efficiency.

[1]  Yuh-Jye Lee,et al.  SSVM: A Smooth Support Vector Machine for Classification , 2001, Comput. Optim. Appl..

[2]  Jaideep Vaidya,et al.  Knowledge and Information Systems , 2007 .

[3]  Stefan Katzenbeisser,et al.  Efficiently Outsourcing Multiparty Computation Under Multiple Keys , 2013, IEEE Transactions on Information Forensics and Security.

[4]  Mingwu Zhang,et al.  On the Soundness and Security of Privacy-Preserving SVM for Outsourcing Data Classification , 2018, IEEE Transactions on Dependable and Secure Computing.

[5]  Taneli Mielikäinen,et al.  Cryptographically private support vector machines , 2006, KDD '06.

[6]  Vinod Vaikuntanathan,et al.  On-the-fly multiparty computation on the cloud via multikey fully homomorphic encryption , 2012, STOC '12.

[7]  Jian Shen,et al.  Privacy preserving multi-party computation delegation for deep learning in cloud computing , 2018, Inf. Sci..

[8]  Matt Blaze,et al.  Divertible Protocols and Atomic Proxy Cryptography , 1998, EUROCRYPT.

[9]  Dan Boneh,et al.  Evaluating 2-DNF Formulas on Ciphertexts , 2005, TCC.

[10]  Yucel Saygin,et al.  Secret charing vs. encryption-based techniques for privacy preserving data mining , 2007 .

[11]  Hitoshi Kiya,et al.  Privacy-Preserving SVM Computing in the Encrypted Domain , 2018, 2018 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC).

[12]  Ming-Syan Chen,et al.  Secure support vector machines outsourcing with random linear transformation , 2014, Knowledge and Information Systems.

[13]  Wei Zhang,et al.  Secure Scalar Product for Big-Data in MapReduce , 2015, 2015 IEEE First International Conference on Big Data Computing Service and Applications.

[14]  Ming Li,et al.  A tale of two clouds: Computing on data encrypted under multiple keys , 2014, 2014 IEEE Conference on Communications and Network Security.

[15]  Ming-Syan Chen,et al.  Privacy-preserving outsourcing support vector machines with random transformation , 2010, KDD.

[16]  John E. Dennis,et al.  Numerical methods for unconstrained optimization and nonlinear equations , 1983, Prentice Hall series in computational mathematics.

[17]  Stratis Ioannidis,et al.  Privacy-Preserving Ridge Regression on Hundreds of Millions of Records , 2013, 2013 IEEE Symposium on Security and Privacy.

[18]  Glenn Fung,et al.  Privacy-preserving classification of vertically partitioned data via random kernels , 2008, TKDD.

[19]  Oded Goldreich,et al.  Foundations of Cryptography: Volume 2, Basic Applications , 2004 .

[20]  Ming Li,et al.  Toward Practical Privacy-Preserving Frequent Itemset Mining on Encrypted Cloud Data , 2020, IEEE Transactions on Cloud Computing.

[21]  Jian Wang,et al.  Secure Naïve Bayesian Classification over Encrypted Data in Cloud , 2016, ProvSec.

[22]  Baocang Wang,et al.  Highly Secure Privacy-Preserving Outsourced k-Means Clustering under Multiple Keys in Cloud Computing , 2020, Secur. Commun. Networks.

[23]  Wei Zhang,et al.  Encrypted SVM for Outsourced Data Mining , 2015, 2015 IEEE 8th International Conference on Cloud Computing.

[24]  Chih-Jen Lin,et al.  A study on reduced support vector machines , 2003, IEEE Trans. Neural Networks.

[25]  Hassan Takabi,et al.  Deep Neural Networks Classification over Encrypted Data , 2019, CODASPY.

[26]  Su-Yun Huang,et al.  Reduced Support Vector Machines: A Statistical Theory , 2007, IEEE Transactions on Neural Networks.

[27]  Artak Amirbekyan,et al.  A New Efficient Privacy-Preserving Scalar Product Protocol , 2007, AusDM.

[28]  Jinshu Su,et al.  Towards Practical Privacy-Preserving Decision Tree Training and Evaluation in the Cloud , 2020, IEEE Transactions on Information Forensics and Security.

[29]  Zoe L. Jiang,et al.  Secure Dot Product of Outsourced Encrypted Vectors and its Application to SVM , 2017, SCC@AsiaCCS.

[30]  Michael Zohner,et al.  ABY - A Framework for Efficient Mixed-Protocol Secure Two-Party Computation , 2015, NDSS.