### Abstract

Social network analysis presupposes that observed social behavior is influenced by an unobserved network. Traditional approaches to inferring the latent network use pairwise descriptive statistics that rely on a variety of measures of co-occurrence. While these techniques have proven useful in a wide range of applications, the literature does not describe the generating mechanism of the observed data from the network. In a previous article, the authors presented a technique which used a finite mixture model as the connection between the unobserved network and the observed social behavior. This model assumed that each group was the result of a star graph on a subset of the population. Thus, each group was the result of a leader who selected members of the population to be in the group. They called these hub models. This approach treats the network values as parameters of a model. However, this leads to a general challenge in estimating parameters which must be addressed. For small datasets there can be far more parameters to estimate than there are observations. Under these conditions, the estimated network can be unstable. In this article, we propose a solution which penalizes the number of nodes which can exert a leadership role. We implement this as a pseudo-Expectation Maximization algorithm. We demonstrate this technique through a series of simulations which show that when the number of leaders is sparse, parameter estimation is improved. Further, we apply this technique to a dataset of animal behavior and an example of recommender systems.

Original language | English (US) |
---|---|

Pages (from-to) | 27-36 |

Number of pages | 10 |

Journal | Social Networks |

Volume | 49 |

DOIs | |

State | Published - May 1 2017 |

Externally published | Yes |

### Fingerprint

### Keywords

- Finite mixture model
- Hub model
- Regularization method
- Social network analysis

### ASJC Scopus subject areas

- Anthropology
- Sociology and Political Science
- Social Sciences(all)
- Psychology(all)

### Cite this

*Social Networks*,

*49*, 27-36. https://doi.org/10.1016/j.socnet.2016.09.003

**Penalized component hub models.** / Weko, Charles; Zhao, Yunpeng.

Research output: Contribution to journal › Article

*Social Networks*, vol. 49, pp. 27-36. https://doi.org/10.1016/j.socnet.2016.09.003

}

TY - JOUR

T1 - Penalized component hub models

AU - Weko, Charles

AU - Zhao, Yunpeng

PY - 2017/5/1

Y1 - 2017/5/1

N2 - Social network analysis presupposes that observed social behavior is influenced by an unobserved network. Traditional approaches to inferring the latent network use pairwise descriptive statistics that rely on a variety of measures of co-occurrence. While these techniques have proven useful in a wide range of applications, the literature does not describe the generating mechanism of the observed data from the network. In a previous article, the authors presented a technique which used a finite mixture model as the connection between the unobserved network and the observed social behavior. This model assumed that each group was the result of a star graph on a subset of the population. Thus, each group was the result of a leader who selected members of the population to be in the group. They called these hub models. This approach treats the network values as parameters of a model. However, this leads to a general challenge in estimating parameters which must be addressed. For small datasets there can be far more parameters to estimate than there are observations. Under these conditions, the estimated network can be unstable. In this article, we propose a solution which penalizes the number of nodes which can exert a leadership role. We implement this as a pseudo-Expectation Maximization algorithm. We demonstrate this technique through a series of simulations which show that when the number of leaders is sparse, parameter estimation is improved. Further, we apply this technique to a dataset of animal behavior and an example of recommender systems.

AB - Social network analysis presupposes that observed social behavior is influenced by an unobserved network. Traditional approaches to inferring the latent network use pairwise descriptive statistics that rely on a variety of measures of co-occurrence. While these techniques have proven useful in a wide range of applications, the literature does not describe the generating mechanism of the observed data from the network. In a previous article, the authors presented a technique which used a finite mixture model as the connection between the unobserved network and the observed social behavior. This model assumed that each group was the result of a star graph on a subset of the population. Thus, each group was the result of a leader who selected members of the population to be in the group. They called these hub models. This approach treats the network values as parameters of a model. However, this leads to a general challenge in estimating parameters which must be addressed. For small datasets there can be far more parameters to estimate than there are observations. Under these conditions, the estimated network can be unstable. In this article, we propose a solution which penalizes the number of nodes which can exert a leadership role. We implement this as a pseudo-Expectation Maximization algorithm. We demonstrate this technique through a series of simulations which show that when the number of leaders is sparse, parameter estimation is improved. Further, we apply this technique to a dataset of animal behavior and an example of recommender systems.

KW - Finite mixture model

KW - Hub model

KW - Regularization method

KW - Social network analysis

UR - http://www.scopus.com/inward/record.url?scp=85000922978&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85000922978&partnerID=8YFLogxK

U2 - 10.1016/j.socnet.2016.09.003

DO - 10.1016/j.socnet.2016.09.003

M3 - Article

AN - SCOPUS:85000922978

VL - 49

SP - 27

EP - 36

JO - Social Networks

JF - Social Networks

SN - 0378-8733

ER -