Network Estimation via Graphon With Node Features Academic Article

abstract

• Estimating the probabilities of linkages in a network has gained increasing interest in recent years. One popular model for network analysis is the exchangeable graph model (ExGM) characterized by a two-dimensional function known as a graphon. Estimating an underlying graphon becomes the key of such analysis. Several nonparametric estimation methods have been proposed, and some are provably consistent. However, if certain useful features of the nodes (e.g., age and schools in social network context) are available, none of these methods was designed to incorporate this source of information to help with the estimation. This paper develops a consistent graphon estimation method that integrates the information from both the adjacency matrix itself and node features. We show that properly leveraging the features can improve the estimation. A cross-validation method is proposed to automatically select the tuning parameter of the method.

authors

• Wong, Raymond Ka Wai

published proceedings

• IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING

author list (cited authors)

• Su, Y. i., Wong, R., & Lee, T.

citation count

• 4

complete list of authors

• Su, Yi||Wong, Raymond KW||Lee, Thomas CM

publication date

• July 2020

publisher

• Institute of Electrical and Electronics Engineers (IEEE)  Publisher

keywords

• Consistency
• Convergence
• Estimation
• Exchangeable Graph Model
• Feature Assisted Neighborhood Smoothing (fans)
• Generative Model
• Nist
• Nonparametric
• Smoothing Methods
• Social Networking (online)
• Synthetic Aperture Sonar
• Tuning

Digital Object Identifier (DOI)

• 10.1109/TNSE.2020.2973994

start page

• 2078

end page

• 2089

volume

• 7

issue

• 3

URL

• http://dx.doi.org/10.1109/tnse.2020.2973994