Link Prediction via Matrix Factorization
Abstract
We propose to solve the link prediction problem in graphs using a supervised matrix factorization approach. The model learns latent features from the topological structure of a (possibly directed) graph, and is shown to make better predictions than popular unsupervised scores. We show how these latent features may be combined with optional explicit features for nodes or edges, which yields better performance than using either type of feature exclusively. Finally, we propose a novel approach to address the class imbalance problem which is common in link prediction by directly optimizing for a ranking loss. Our model is optimized with stochastic gradient descent and scales to large graphs. Results on several datasets show the efficacy of our approach.
Cite
Text
Menon and Elkan. "Link Prediction via Matrix Factorization." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2011. doi:10.1007/978-3-642-23783-6_28Markdown
[Menon and Elkan. "Link Prediction via Matrix Factorization." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2011.](https://mlanthology.org/ecmlpkdd/2011/menon2011ecmlpkdd-link/) doi:10.1007/978-3-642-23783-6_28BibTeX
@inproceedings{menon2011ecmlpkdd-link,
title = {{Link Prediction via Matrix Factorization}},
author = {Menon, Aditya Krishna and Elkan, Charles},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2011},
pages = {437-452},
doi = {10.1007/978-3-642-23783-6_28},
url = {https://mlanthology.org/ecmlpkdd/2011/menon2011ecmlpkdd-link/}
}