Article: A new non-parametric feature learning for supervised link prediction Journal: International Journal of System Control and Information Processing (IJSCIP) 2015 Vol.1 No.4 pp.319 - 339 Abstract: Link prediction is an important task for analysing relational data such as the friendship relation on a social networking website that also has applications in other domains like, information retrieval, bioinformatics and e-commerce. The problem of link prediction is to predict the existence or absence of edges between vertices of a network. In this paper, we present a novel non-parametric latent feature relational model based on distance dependent Indian buffet process (DDIBP), by which we can utilise the information of topological structure of the network such as shortest path and connectivity of the nodes and incorporate them into the proposed Bayesian Non-parametric latent feature model which can automatically infer the unknown latent feature dimension. We also develop an efficient MCMC algorithm to compute the posterior distribution of the hidden variables with a highly nonlinear link likelihood function. Experimental results on four real datasets demonstrate the superiority of the proposed method over other latent feature models for link prediction problem. Inderscience Publishers - linking academia, business and industry through research

Title: A new non-parametric feature learning for supervised link prediction

Authors: Ahmad Agha Kardan; Samira Ghareh Gozlou

Addresses: Department of Computer Engineering and Information Technology, Amirkabir University of Technology (Tehran Polytechnic), Tehran 11155-11365, Iran ' Department of Computer Engineering and Information Technology, Amirkabir University of Technology (Tehran Polytechnic), Tehran 11155-11365, Iran

Abstract: Link prediction is an important task for analysing relational data such as the friendship relation on a social networking website that also has applications in other domains like, information retrieval, bioinformatics and e-commerce. The problem of link prediction is to predict the existence or absence of edges between vertices of a network. In this paper, we present a novel non-parametric latent feature relational model based on distance dependent Indian buffet process (DDIBP), by which we can utilise the information of topological structure of the network such as shortest path and connectivity of the nodes and incorporate them into the proposed Bayesian Non-parametric latent feature model which can automatically infer the unknown latent feature dimension. We also develop an efficient MCMC algorithm to compute the posterior distribution of the hidden variables with a highly nonlinear link likelihood function. Experimental results on four real datasets demonstrate the superiority of the proposed method over other latent feature models for link prediction problem.

Keywords: supervised link prediction; latent feature models; DDIBP; distance dependent Indian buffet process; Gibbs sampling; convex duality; information processing; nonparametric feature learning; relational modelling; topological structure; network structure; shortest path; node connectivity; network topology.

DOI: 10.1504/IJSCIP.2015.075877

International Journal of System Control and Information Processing, 2015 Vol.1 No.4, pp.319 - 339

Received: 03 Feb 2014
Accepted: 13 Jun 2014
Published online: 12 Apr 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: A new non-parametric feature learning for supervised link prediction

Keep up-to-date