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:
|
229
|
Type:
|
Topic Contributed
|
Date/Time:
|
Monday, July 30, 2012 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Health Policy Statistics Section
|
Abstract - #305444 |
Title:
|
Modeling Latent Tie Strength in Social Networks Using Interaction Count Data
|
Author(s):
|
Mauricio Sadinle*+
|
Companies:
|
Carnegie Mellon University
|
Address:
|
, Pittsburgh, PA, 15217,
|
Keywords:
|
Online social networks ;
Valued networks ;
Count data ;
Negative binomial model ;
Latent variable model ;
EM algorithm
|
Abstract:
|
We propose a model to measure latent tie strength between members of social networks. The motivation to do so is to have a more reliable measure of relationship between members of the network besides declared links. This methodology has applications to link predictions, item recommendation, newsfeeds, people search and other related tasks implemented in online social networks. The proposed model uses different types of count interaction data between members such as number of messages and chats. The intuition for the model is that the more interaction the larger the latent tie strength. We propose a mixture model that fits marginally the distribution of the observed interaction between users. Our model also allows the incorporation of covariates to explain the mean latent tie strength. We illustrate the usefulness of our model using data from the myGamma mobile social network.
|
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.