Robust graph learning with graph convolutional network

Yingying Wan

doi:10.1016/j.ipm.2022.102916

Robust graph learning with graph convolutional network

2022-05-01

12

PRE

AI

DOI：10.1016/j.ipm.2022.102916

Original

Original request for help

Save

Abstract

En 中文

Graph convolutional network (GCN) is a powerful tool to process the graph data and hasachieved satisfactory performance in the task of node classification. In general, GCN uses a fixedgraph to guide the graph convolutional operation. However, the fixed graph from the originalfeature space may contain noises or outliers, which may degrade the effectiveness of GCN. Toaddress this issue, in this paper, we propose a robust graph learning convolutional network(RGLCN). Specifically, we design a robust graph learning model based on the sparse constraintand strong connectivity constraint to achieve the smoothness of the graph learning. In addition,we introduce graph learning model into GCN to explore the representative information, aimingto learning a high-quality graph for the downstream task. Experiments on citation networkdatasets show that the proposed RGLCN outperforms the existing comparison methods withrespect to the task of node classification

Keywords:

Graph convolutional network

Node classification

Sparse constraint

Robust graph learning with graph convolutional network

Abstract

Journal

Organization