Graph Convolutional Matrix Completion
Rianne van den Berg, Thomas N. Kipf, Max Welling
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/riannevdberg/gc-mcOfficialtf★ 0
- github.com/lmcRS/AWS-recommendation-papersnone★ 60
- github.com/blueghostyi/id-grecpytorch★ 30
- github.com/hengruizhang98/GCMC-Pytorch-dglpytorch★ 28
- github.com/swtheing/Multiview-Link-Representation-Learningpytorch★ 17
- github.com/OweysMomenzada/Graph-Neural-Networks-for-effecient-Recommender-Systemspytorch★ 6
- github.com/tanimutomo/gcmcpytorch★ 0
- github.com/paryanasr/HonoursProject-Recommender-System-gc-mctf★ 0
- github.com/dmlc/dgl/tree/master/examples/mxnet/_deprecated/gcmcpytorch★ 0
- github.com/tjcdev/gcmc-tf2tf★ 0
Abstract
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. Building on recent progress in deep learning on graph-structured data, we propose a graph auto-encoder framework based on differentiable message passing on the bipartite interaction graph. Our model shows competitive performance on standard collaborative filtering benchmarks. In settings where complimentary feature information or structured data such as a social network is available, our framework outperforms recent state-of-the-art methods.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| Douban Monti | GC-MC | RMSE | 0.73 | — | Unverified |
| Flixster Monti | GC-MC | RMSE | 0.92 | — | Unverified |
| MovieLens 100K | GC-MC | RMSE (u1 Splits) | 0.91 | — | Unverified |
| MovieLens 100K | GC-MC | RMSE (u1 Splits) | 0.91 | — | Unverified |
| MovieLens 10M | GC-MC | RMSE | 0.78 | — | Unverified |
| MovieLens 1M | GC-MC | RMSE | 0.83 | — | Unverified |
| YahooMusic Monti | GC-MC | RMSE | 20.5 | — | Unverified |