Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Jun;46(3):480-96.
doi: 10.1016/j.jbi.2013.03.008. Epub 2013 Apr 4.

EXpectation Propagation LOgistic REgRession (EXPLORER): distributed privacy-preserving online model learning

Shuang Wang  1 , Xiaoqian Jiang, Yuan Wu, Lijuan Cui, Samuel Cheng, Lucila Ohno-Machado
Affiliations

EXpectation Propagation LOgistic REgRession (EXPLORER): distributed privacy-preserving online model learning

Shuang Wang et al. J Biomed Inform. 2013 Jun.

Abstract

We developed an EXpectation Propagation LOgistic REgRession (EXPLORER) model for distributed privacy-preserving online learning. The proposed framework provides a high level guarantee for protecting sensitive information, since the information exchanged between the server and the client is the encrypted posterior distribution of coefficients. Through experimental results, EXPLORER shows the same performance (e.g., discrimination, calibration, feature selection, etc.) as the traditional frequentist logistic regression model, but provides more flexibility in model updating. That is, EXPLORER can be updated one point at a time rather than having to retrain the entire data set when new observations are recorded. The proposed EXPLORER supports asynchronized communication, which relieves the participants from coordinating with one another, and prevents service breakdown from the absence of participants or interrupted communications.

PubMed Disclaimer

Figures

Figure D.7
Figure D.7
Heterogeneity between 2 different EXPLORER sites for dataset 4 over 30 trials.
Figure D.8
Figure D.8
Heterogeneity between 2 different EXPLORER sites for dataset 5 over 30 trials.
Figure D.9
Figure D.9
Heterogeneity between 2 different EXPLORER sites for dataset 6 over 30 trials.
Figure D.10
Figure D.10
Heterogeneity between 2 different EXPLORER sites for dataset 7 over 30 trials.
Figure D.11
Figure D.11
Heterogeneity between 2 different EXPLORER sites for dataset 8 over 30 trials.
Figure D.12
Figure D.12
Heterogeneity between 2 different EXPLORER sites for dataset 9 over 30 trials.
Figure D.13
Figure D.13
Heterogeneity between 2 different EXPLORER sites for dataset 10 over 30 trials.
Figure D.14
Figure D.14
Heterogeneity between 2 different EXPLORER sites for dataset 11 over 30 trials.
Figure 1
Figure 1
Factor graph of EXPLORER with 3-site asynchronous update.
Figure 2
Figure 2
Secured intermediate information exchange (SINE) protocol.
Figure 3
Figure 3
The convergence speed of all 11 coefficients of the Edinburgh dataset for an asynchronous 8-site EXPLORE setup.
Figure 4
Figure 4
The convergence speed of evenly partitioned datasets; a) 4-site setup; b) 8-site setup; c) The MSE of each site after the 1st iteration update for n-site setups with n = 2, 3, · · ·, 8.
Figure 5
Figure 5
The convergence speed of unevenly partitioned dataset; a) site 1 with 500 records; b) site 1 with 750 records; c) site 1 with 1000 records.
Figure D.6
Figure D.6
Heterogeneity between 2 different EXPLORER sites for dataset 3 over 30 trials.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Hosmer D, Lemeshow S. Applied logistic regression. Wiley-Interscience; 2000.
    1. Boxwala A, Kim J, Grillo J, Ohno-Machado L. Using statistical and machine learning to help institutions detect suspicious access to electronic health records. J Am Med Inform Assoc. 2011;18:498–505. - PMC - PubMed
    1. Ohno-Machado L. Modeling medical prognosis: survival analysis techniques. J Biomed Inform. 2001;34:428–439. - PubMed
    1. Kennedy R, Fraser H, McStay L, Harrison R. Early diagnosis of acute myocardial infarction using clinical and electrocardiographic data at presentation: derivation and evaluation of logistic regression models. Eur Heart J. 1996;17:1181–1191. - PubMed
    1. Anderson N, Edwards K. Building a chain of trust: using policy and practice to enhance trustworthy clinical data discovery and sharing. Proceedings of the 2010 Workshop on Governance of Technology, Information and Policies, ACM; Austin, Texas, USA. pp. 15–20.

Publication types