Advanced

Sender- and receiver-specific blockmodels

Geng, Zhi LU and Nowicki, Krzysztof LU (2015) In Journal of Social Structure 16. p.1-34
Abstract
We propose a sender-specific blockmodel for network data which utilizes both the group membership and the identities of the vertices.

This is accomplished by introducing the edge probabilities $(\theta_{i, v})$ for $1\le i\le c, 1\le v\le n$, where $i$ specifies the group membership of a sending vertex and $v$ specifies the identity of the receiving vertex. In addition, group membership is consider to be random, with parameters $(p_i)_{i=1}^c$. We present methods based on the EM algorithm for the parameter estimations and discuss the recovery of latent group memberships. A companion model, the receiver-specific blockmodel, is also introduced in which the edge probabilities $(\psi_{u, j})$ for $1\le u \le n,1\le j\le c$ depend on... (More)
We propose a sender-specific blockmodel for network data which utilizes both the group membership and the identities of the vertices.

This is accomplished by introducing the edge probabilities $(\theta_{i, v})$ for $1\le i\le c, 1\le v\le n$, where $i$ specifies the group membership of a sending vertex and $v$ specifies the identity of the receiving vertex. In addition, group membership is consider to be random, with parameters $(p_i)_{i=1}^c$. We present methods based on the EM algorithm for the parameter estimations and discuss the recovery of latent group memberships. A companion model, the receiver-specific blockmodel, is also introduced in which the edge probabilities $(\psi_{u, j})$ for $1\le u \le n,1\le j\le c$ depend on the membership of a vertex receiving a directed edge. We apply both models to several sets of social network data. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Directed graph, Blockmodeling, Out-nets, In-nets, Ego-nets, EM algorithm, Multinomial distribution
in
Journal of Social Structure
volume
16
pages
1 - 34
publisher
International Network for Social Network Analysis (INSNA)
external identifiers
  • scopus:84947760064
ISSN
1529-1227
language
English
LU publication?
yes
id
4c4dc634-80ff-446f-aa16-51684030dbef (old id 8165926)
alternative location
https://www.cmu.edu/joss/content/articles/volume16/GengNowicki.pdf
date added to LUP
2015-11-05 13:19:26
date last changed
2017-01-01 06:26:30
@article{4c4dc634-80ff-446f-aa16-51684030dbef,
  abstract     = {We propose a sender-specific blockmodel for network data which utilizes both the group membership and the identities of the vertices.<br/><br>
This is accomplished by introducing the edge probabilities $(\theta_{i, v})$ for $1\le i\le c, 1\le v\le n$, where $i$ specifies the group membership of a sending vertex and $v$ specifies the identity of the receiving vertex. In addition, group membership is consider to be random, with parameters $(p_i)_{i=1}^c$. We present methods based on the EM algorithm for the parameter estimations and discuss the recovery of latent group memberships. A companion model, the receiver-specific blockmodel, is also introduced in which the edge probabilities $(\psi_{u, j})$ for $1\le u \le n,1\le j\le c$ depend on the membership of a vertex receiving a directed edge. We apply both models to several sets of social network data.},
  author       = {Geng, Zhi and Nowicki, Krzysztof},
  issn         = {1529-1227},
  keyword      = {Directed graph,Blockmodeling,Out-nets,In-nets,Ego-nets,EM algorithm,Multinomial distribution},
  language     = {eng},
  pages        = {1--34},
  publisher    = {International Network for Social Network Analysis (INSNA)},
  series       = {Journal of Social Structure},
  title        = {Sender- and receiver-specific blockmodels},
  volume       = {16},
  year         = {2015},
}