Metric learning-enhanced semi-supervised Graph Convolutional Network for multi-view learning

doi:10.1016/j.inffus.2025.103420

Metric learning-enhanced semi-supervised Graph Convolutional Network for multi-view learning

2025-06-23

0

OA

AI

DOI：10.1016/j.inffus.2025.103420

Original

Original request for help

Save

Abstract

En 中文

• We tackle deep graph-based semi-supervised classification for multi-view data. • In each view, both a KNN graph and a semi-supervised graph are reconstructed. • We devise a network of GCNs that use individual graphs as well as the fused graph. • Metric learning on hidden features utilizes semi-hard examples within a triplet loss. • Experiments are conducted on six public multi-view datasets.

Keywords:

Information fusion with deep learning

Multi-view learning

Semi-supervised classification

Metric learning

Graph Convolutional Network

Metric learning-enhanced semi-supervised Graph Convolutional Network for multi-view learning

Abstract

Journal

Organization