The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
391
|
Type:
|
Topic Contributed
|
Date/Time:
|
Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
|
Sponsor:
|
ENAR
|
Abstract - #305594 |
Title:
|
Link Prediction for Partially Observed Networks
|
Author(s):
|
Yunpeng Zhao*+ and Liza Levina and Ji Zhu
|
Companies:
|
University of Michigan and University of Michigan and University of Michigan
|
Address:
|
, , ,
|
Keywords:
|
link prediction ;
supervised learning
|
Abstract:
|
A fundamental problem in network analysis is link prediction for partially observed networks. One difficulty of applying existing supervised learning approaches to link prediction is a lack of negative examples, i.e., there is often no certain information that no link exists between a pair. To address this situation, we present a new method, which treats all observed links as true positives and others as missing data. The missing rate cannot be estimated in this setup, but a relative ranking of probabilities of potential links can be. Our method provides a ranked list of pairs ordered by probability of link, which is sufficient in many applications. To obtain these rankings, we utilize information on node covariates as well as network topology. The method performs well on both simulated and real-world networks.
|
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 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.