Bipartite Edge Prediction via Transductive Learning over Product Graphs

Hanxiao Liu, Yiming Yang
Proceedings of the 32nd International Conference on Machine Learning, PMLR 37:1880-1888, 2015.

Abstract

This paper addresses the problem of predicting the missing edges of a bipartite graph where each side of the vertices has its own intrinsic structure. We propose a new optimization framework to map the two sides of the intrinsic structures onto the manifold structure of the edges via a graph product, and to reduce the original problem to vertex label propagation over the product graph. This framework enjoys flexible choices in the formulation of graph products, and supports a rich family of graph transduction schemes with scalable inference. Experiments on benchmark datasets for collaborative filtering, citation network analysis and prerequisite prediction of online courses show advantageous performance of the proposed approach over other state-of-the-art methods.

Cite this Paper


BibTeX
@InProceedings{pmlr-v37-liuc15, title = {Bipartite Edge Prediction via Transductive Learning over Product Graphs}, author = {Liu, Hanxiao and Yang, Yiming}, booktitle = {Proceedings of the 32nd International Conference on Machine Learning}, pages = {1880--1888}, year = {2015}, editor = {Bach, Francis and Blei, David}, volume = {37}, series = {Proceedings of Machine Learning Research}, address = {Lille, France}, month = {07--09 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v37/liuc15.pdf}, url = {https://proceedings.mlr.press/v37/liuc15.html}, abstract = {This paper addresses the problem of predicting the missing edges of a bipartite graph where each side of the vertices has its own intrinsic structure. We propose a new optimization framework to map the two sides of the intrinsic structures onto the manifold structure of the edges via a graph product, and to reduce the original problem to vertex label propagation over the product graph. This framework enjoys flexible choices in the formulation of graph products, and supports a rich family of graph transduction schemes with scalable inference. Experiments on benchmark datasets for collaborative filtering, citation network analysis and prerequisite prediction of online courses show advantageous performance of the proposed approach over other state-of-the-art methods.} }
Endnote
%0 Conference Paper %T Bipartite Edge Prediction via Transductive Learning over Product Graphs %A Hanxiao Liu %A Yiming Yang %B Proceedings of the 32nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2015 %E Francis Bach %E David Blei %F pmlr-v37-liuc15 %I PMLR %P 1880--1888 %U https://proceedings.mlr.press/v37/liuc15.html %V 37 %X This paper addresses the problem of predicting the missing edges of a bipartite graph where each side of the vertices has its own intrinsic structure. We propose a new optimization framework to map the two sides of the intrinsic structures onto the manifold structure of the edges via a graph product, and to reduce the original problem to vertex label propagation over the product graph. This framework enjoys flexible choices in the formulation of graph products, and supports a rich family of graph transduction schemes with scalable inference. Experiments on benchmark datasets for collaborative filtering, citation network analysis and prerequisite prediction of online courses show advantageous performance of the proposed approach over other state-of-the-art methods.
RIS
TY - CPAPER TI - Bipartite Edge Prediction via Transductive Learning over Product Graphs AU - Hanxiao Liu AU - Yiming Yang BT - Proceedings of the 32nd International Conference on Machine Learning DA - 2015/06/01 ED - Francis Bach ED - David Blei ID - pmlr-v37-liuc15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 37 SP - 1880 EP - 1888 L1 - http://proceedings.mlr.press/v37/liuc15.pdf UR - https://proceedings.mlr.press/v37/liuc15.html AB - This paper addresses the problem of predicting the missing edges of a bipartite graph where each side of the vertices has its own intrinsic structure. We propose a new optimization framework to map the two sides of the intrinsic structures onto the manifold structure of the edges via a graph product, and to reduce the original problem to vertex label propagation over the product graph. This framework enjoys flexible choices in the formulation of graph products, and supports a rich family of graph transduction schemes with scalable inference. Experiments on benchmark datasets for collaborative filtering, citation network analysis and prerequisite prediction of online courses show advantageous performance of the proposed approach over other state-of-the-art methods. ER -
APA
Liu, H. & Yang, Y.. (2015). Bipartite Edge Prediction via Transductive Learning over Product Graphs. Proceedings of the 32nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 37:1880-1888 Available from https://proceedings.mlr.press/v37/liuc15.html.

Related Material