|
Activity Number:
|
191
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #308792 |
|
Title:
|
Secure Logistic Regression
|
|
Author(s):
|
Yuval Nardi*+ and Stephen Fienberg and Aleksandra Slavkovic
|
|
Companies:
|
Carnegie Mellon University and Carnegie Mellon University and The Pennsylvania State University
|
|
Address:
|
5921 Douglas St, Pittsburgh, PA, 15217,
|
|
Keywords:
|
|
|
Abstract:
|
We address the problem of performing a logistic regression in a secure way without directly sharing data among parties. We suppose that data are collected separately and intimately by several separate parties (agencies). These parties wish to analyze the pooled (combined) data without actually combining the parts that they possess (i.e., they want to fit a model and make inferences using the pooled data in a way that no one party's data are disclosed to another party). In this paper we build on earlier results by Fienberg, Fulp, Slavkovic and Wrobel (2006) on the horizontally partitioned case and describe methods for both the vertically partitioned case and the general case, the vertically partitioned, partially overlapping data.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2007 program |
