|
Activity Number:
|
317
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #301819 |
|
Title:
|
Methods for Time-Dependent Network Analysis
|
|
Author(s):
|
Andrew C. Thomas*+ and Joseph Blitzstein
|
|
Companies:
|
Harvard University and Harvard University
|
|
Address:
|
One Oxford Street, 7th Floor, Cambridge, MA, 02138,
|
|
Keywords:
|
social ; networks ; hypergraph
|
|
Abstract:
|
We illustrate the extension of methods for network analysis beyond a static link approach, as commonly seen in social network analysis, and a dyadic approach, as seen in most types of network analyses, to a more general dynamic-linked higher-order graph system. We demonstrate these methods on a long-term data set of dynamic social connections for the inference and prediction of social patterns.
|
