Secure Regression for Vertically Partitioned, Partially Overlapping Data
نویسندگان
چکیده
We consider the setting where multiple parties with different variables and units seek to combine their data to fit regressions but are not willing or not allowed to share their data values. We present a general strategy to tackle such problems by treating them as missing data problems, and we estimate regression coefficients using secure EM algorithms. We present secure EM algorithms for linear and log-linear regressions, based on the multivariate normal and multinomial distributions. The parties compute and share the sufficient statistics required for the EM algorithms via secure matrix product protocols, which avoid sharing individual data values.
منابع مشابه
Privacy-Preserving Analysis of Vertically Partitioned Data Using Secure Matrix Products
Reluctance of statistical agencies and other data owners to share their possibly confidential or proprietary data with others who own related databases is a serious impediment to conducting mutually beneficial analyses. In this paper, we propose a protocol for securely computing matrix products in vertically partitioned data, i.e., the data sets have the same subjects but disjoint attributes. T...
متن کاملPrivacy Preserving Data Mining over Vertically Partitioned Data
Vaidya, Jaideep Shrikant. Ph.D., Purdue University, August, 2004. Privacy Preserving Data Mining over Vertically Partitioned Data. Major Professor: Chris Clifton. The goal of data mining is to extract or “mine” knowledge from large amounts of data. However, data is often collected by several different sites. Privacy, legal and commercial concerns restrict centralized access to this data. Theore...
متن کاملSMC Protocol for Naïve Bayes Classification over Grid Partitioned Data using Multiple UTPs
The case where data is distributed horizontally as well as vertically, it refers as grid partitioned data. SMC protocol for Naïve Bayes classification over grid partitioned data is offered in this paper. Also present a solution of the Secure Multi-party Computation (SMC) problem in the form of a protocol that preserves privacy. In this system, a protocol with several Un-trusted Third Parties (U...
متن کاملImplementing Privacy-Preserving Bayesian-Net Discovery for Vertically Partitioned Data
The great potential of data mining in a networked world cannot be realized without acceptable guarantees that private information will be protected. In theory, general cryptographic protocols for secure multiparty computation enable data mining with privacy preservation that is optimal with respect to the desired end results. However, the performance expense of such general protocols is prohibi...
متن کاملFast Steganography-based Multi-Party Protocols for Privacy-Preserving Association Rule Mining in Vertically Partitioned Data
Recently, with the emergence of privacy issues in data mining, considerable research has focused on developing new data mining algorithms that incorporate privacy constraints, and, in the same time, are as efficient as possible in terms of accuracy of the results. In this paper, we focus on privately mining association rules in vertically partitioned data, and propose two steganography-based mu...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2004