Leveraging Node Attributes for Incomplete Relational Data

He Zhao, Lan Du, Wray Buntine
Proceedings of the 34th International Conference on Machine Learning, PMLR 70:4072-4081, 2017.

Abstract

Relational data are usually highly incomplete in practice, which inspires us to leverage side information to improve the performance of community detection and link prediction. This paper presents a Bayesian probabilistic approach that incorporates various kinds of node attributes encoded in binary form in relational models with Poisson likelihood. Our method works flexibly with both directed and undirected relational networks. The inference can be done by efficient Gibbs sampling which leverages sparsity of both networks and node attributes. Extensive experiments show that our models achieve the state-of-the-art link prediction results, especially with highly incomplete relational data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v70-zhao17a, title = {Leveraging Node Attributes for Incomplete Relational Data}, author = {He Zhao and Lan Du and Wray Buntine}, booktitle = {Proceedings of the 34th International Conference on Machine Learning}, pages = {4072--4081}, year = {2017}, editor = {Precup, Doina and Teh, Yee Whye}, volume = {70}, series = {Proceedings of Machine Learning Research}, month = {06--11 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v70/zhao17a/zhao17a.pdf}, url = {https://proceedings.mlr.press/v70/zhao17a.html}, abstract = {Relational data are usually highly incomplete in practice, which inspires us to leverage side information to improve the performance of community detection and link prediction. This paper presents a Bayesian probabilistic approach that incorporates various kinds of node attributes encoded in binary form in relational models with Poisson likelihood. Our method works flexibly with both directed and undirected relational networks. The inference can be done by efficient Gibbs sampling which leverages sparsity of both networks and node attributes. Extensive experiments show that our models achieve the state-of-the-art link prediction results, especially with highly incomplete relational data.} }
Endnote
%0 Conference Paper %T Leveraging Node Attributes for Incomplete Relational Data %A He Zhao %A Lan Du %A Wray Buntine %B Proceedings of the 34th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2017 %E Doina Precup %E Yee Whye Teh %F pmlr-v70-zhao17a %I PMLR %P 4072--4081 %U https://proceedings.mlr.press/v70/zhao17a.html %V 70 %X Relational data are usually highly incomplete in practice, which inspires us to leverage side information to improve the performance of community detection and link prediction. This paper presents a Bayesian probabilistic approach that incorporates various kinds of node attributes encoded in binary form in relational models with Poisson likelihood. Our method works flexibly with both directed and undirected relational networks. The inference can be done by efficient Gibbs sampling which leverages sparsity of both networks and node attributes. Extensive experiments show that our models achieve the state-of-the-art link prediction results, especially with highly incomplete relational data.
APA
Zhao, H., Du, L. & Buntine, W.. (2017). Leveraging Node Attributes for Incomplete Relational Data. Proceedings of the 34th International Conference on Machine Learning, in Proceedings of Machine Learning Research 70:4072-4081 Available from https://proceedings.mlr.press/v70/zhao17a.html.

Related Material