New Graph Neural Network Method Improves Multi-view Learning by Better Leveraging Data Consistency Across Multiple Perspectives

Researchers propose a Multi-view Graph Convolutional Network that addresses limitations in existing GCN-based approaches for multi-view data analysis

The method replaces traditional KNN topology construction with granular-ball-based topology to avoid artificial k-value constraints and better capture inter-node consistency

The approach enhances feature quality by addressing overlooked inter-feature consistency within individual views, improving final embedding representations

The model implements interactive fusion of multiple views during the graph convolution process rather than after, enabling fuller utilization of inter-view consistency