Graph Embedded Contrastive Learning for Multi-View Clustering

Multi-View Clustering (MVC) and Multi-View Graph Clustering (MVGC) are often studied separately, which prevents the advantages of these two approaches from being fully utilized and limits their application domains. This project contains the code for the IJCAI 2025 paper GMVC, where we expect to enable a unified perspective in addressing the subproblems of MVC and MVGC, as well as to reduce the repetitive labor that might be avoidable between MVC and MVGC for the future.

Installation

• pip install -r requirements.txt
• Download the datasets
• Download the trained models
• The complete documents for reproduction are provided in Download.

To test the trained model, run:

python test.py

To train a new model, run:

python main_gmvc.py