Graph Convolutional Matrix Completion

7 Jun 2017Rianne van den BergThomas N. KipfMax Welling

We consider matrix completion for recommender systems from the point of view of link prediction on graphs. Interaction data such as movie ratings can be represented by a bipartite user-item graph with labeled edges denoting observed ratings... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Recommendation Systems Douban GC-MC RMSE 0.734 # 4
Recommendation Systems Flixster GC-MC RMSE 0.917 # 3
Recommendation Systems MovieLens 100K GC-MC RMSE 0.910 # 2
Recommendation Systems MovieLens 100K GC-MC + feat RMSE 0.905 # 1
Recommendation Systems MovieLens 10M GC-MC RMSE 0.777 # 11
Recommendation Systems MovieLens 1M GC-MC RMSE 0.832 # 4
Recommendation Systems YahooMusic GC-MC RMSE 20.5 # 2