Secure multi-party linear regression

Fida Dankar, Renaud Brien, Carlisle Adams, Stan Matwin

Research output: Contribution to journalConference articlepeer-review

12 Citations (Scopus)

Abstract

Increasing efficiency in hospitals is of particular importance. Studies that combine data from multiple hospitals/data holders can tremendously improve the statistical outcome and aid in identifying efficiency markers. However, combining data from multiple sources for analysis poses privacy risks. A number of protocols have been proposed in the literature to address the privacy concerns; however they do not fully deliver on either privacy or complexity. In this paper, we present a privacy preserving linear regression model for the analysis of data coming from several sources. The protocol uses a semi-trusted third party and delivers on privacy and complexity.

Original languageEnglish
Pages (from-to)406-414
Number of pages9
JournalCEUR Workshop Proceedings
Volume1133
Publication statusPublished - 2014
Externally publishedYes
Event2014 Joint Workshops on International Conference on Extending Database Technology, EDBT 2014 and International Conference on Database Theory, ICDT 2014 - Athens, Greece
Duration: Mar 28 2014 → …

Keywords

  • Linear regression
  • Privacy preserving data mining
  • Secure multiparty computation

ASJC Scopus subject areas

  • General Computer Science

Fingerprint

Dive into the research topics of 'Secure multi-party linear regression'. Together they form a unique fingerprint.

Cite this