Fast link prediction for large networks using spectral embedding
Social and Information Networks
2017-04-10 v1 Physics and Society
Abstract
Many link prediction algorithms require the computation of a similarity metric on each vertex pair, which is quadratic in the number of vertices and infeasible for large networks. We develop a class of link prediction algorithms based on a spectral embedding and the k closest pairs algorithm that are scalable to very large networks. We compare the prediction accuracy and runtime of these methods to existing algorithms on several large link prediction tasks. Our methods achieve comparable accuracy to standard algorithms but are significantly faster.
Cite
@article{arxiv.1703.09693,
title = {Fast link prediction for large networks using spectral embedding},
author = {Benjamin Pachev and Benjamin Webb},
journal= {arXiv preprint arXiv:1703.09693},
year = {2017}
}