Few-shot node classification via local adaptive discriminant structure learning

• Published in: Frontiers of Computer Science Vol. 17; no. 2; p. 172316
• Main Authors: XUE, Zhe, DU, Junping, XU, Xin, LIU, Xiangbin, WANG, Junfu, KOU, Feifei
• Format: Journal Article
• Language: English
• Published: Beijing Higher Education Press 01.04.2023; Springer Nature B.V
• Subjects: adaptive structure learning; Algorithms and Applications; attention strategy; Citation analysis; Classification; Computer Science; few-shot learning; graph neural network; Learning; Metric space; Network analysis; Neural networks; node classification; Nodes; Optimization; Research Article; Social networks; Special Section on Meta-learning: Theories; Teaching methods; node classification; adaptive structure learning; attention strategy; graph neural network; few-shot learning
• ISSN: 2095-2228; 2095-2236
• DOI: 10.1007/s11704-022-1259-6

Node classification has a wide range of application scenarios such as citation analysis and social network analysis. In many real-world attributed networks, a large portion of classes only contain limited labeled nodes. Most of the existing node classification methods cannot be used for few-shot node classification. To train the model effectively and improve the robustness and reliability of the model with scarce labeled samples, in this paper, we propose a local adaptive discriminant structure learning (LADSL) method for few-shot node classification. LADSL aims to properly represent the nodes in the attributed graphs and learn a metric space with a strong discriminating power by reducing the intra-class variations and enlarging inter-class differences. Extensive experiments conducted on various attributed networks datasets demonstrate that LADSL is superior to the other methods on few-shot node classification task.