全部 |
  • 全部
  • 题名
  • 作者
  • 机构
  • 关键词
  • NSTL主题词
  • 摘要
检索 二次检索 AI检索
外文文献 中文文献
筛选条件:

1. Motif-aware curriculum learning for node classification NSTL国家科技图书文献中心

Cai, Xiaosha |  Chen, Man-Sheng... -  《Neural Networks》 - 2025,184 - Article 107089~Article 107089 - 共13页

摘要:Node classification, seeking to predict the |  learning-based node classification methods fail to |  Curriculum Learning for Node Classification (MACL). It |  learning. One of the most popular methods for node |  classification is currently Graph Neural Networks (GNNs
关键词: Node classification |  Curriculum learning |  Motif-aware |  Subgraph information

2. Data-free Knowledge Distillation based on GNN for Node Classification NSTL国家科技图书文献中心

Xinfeng Zeng |  Tao Liu... -  《Database Systems for Advanced Applications,Part II》 -  International Conference on Database Systems for Advanced Applications |  International Workshop on Big Data Management and Service |  International Workshop on Graph Data Management and Analysis |  International Workshop on Big Data Quality Management |  Workshop on Emerging Results inData Science and Engineering - 2025, - 243~258 - 共16页

摘要: Knowledge Distillation framework for Node Classification |  in node classification tasks. Code is available at |  methods are primarily designed for graph classification |  tasks involving small-sized graphs. Research on node |  classification tasks for larger graphs remains unexplored. To
关键词: Graph neural networks |  Data-Free knowledge distillation |  Node classification

3. Community-Hop: Enhancing Node Classification Through Community Preference NSTL国家科技图书文献中心

Ahmed Begga |  Waqar Ali... -  《Structural,Syntactic,and Statistical Pattern Recognition》 -  IAPR International Workshops on Structural and Syntactic Pattern Recognition |  Statistical Techniques in Pattern Recognition - 2025, - 21~30 - 共10页

摘要: and perform various graph-related tasks from node |  classification to link prediction. Recently, GNNs have mostly |  ability to combine node features and graph topology |  improve performance on node-level tasks. Extensive |  experiments on six node-level datasets under standard
关键词: Graph neural networks |  Node classification |  Spectral clustering

4. Label as Equilibrium: A performance booster for Graph Neural Networks on node classification NSTL国家科技图书文献中心

Luo, Yi |  Luo, Guangchun... -  《Neural Networks》 - 2025,186 - Article 107284~Article 107284 - 共12页

摘要: node classification task. Recently, a series of label |  reuse approaches emerged to boost the node |  classification performance of GNN. They repeatedly input the |  predicted node class labels into the underlying GNN to |  constant memory consumption. Excessive node
关键词: Graph Neural Networks |  Node classification

5. SLRNode: node similarity-based leading relationship representation layer in graph neural networks for node classification NSTL国家科技图书文献中心

Fuchuan Xiang |  Yao Xiao... -  《The Journal of Supercomputing》 - 2025,81(5) - 657.1~657.34 - 共34页

摘要:For semi-supervised node classification in |  structure into the GNN and introduces a node similarity |  self-supervised manner and added to the node label |  other model is used to predict node labels. Node |  classification experiments were conducted on six datasets to
关键词: Node classification |  GNNs |  Leading tree |  Dual-model pseudo-label training framework

6. GraphDHV: Graph Neural Network with Dual Hybrid View on Imbalanced Node Classification NSTL国家科技图书文献中心

Longqing Du |  Guangquan Lu... -  《Computing and Combinatorics,Part II》 -  International Computing and Combinatorics Conference - 2025, - 502~513 - 共12页

摘要: real-world data on node classification in a graph |  encoder that integrates node attributes and topological | It is crucial to address category imbalance in | . This paper introduces the GraphDHV model, a dual |  information. GraphDHV improves inter-class separability and
关键词: Node classification |  Class imbalance |  Graph neural network

7. Joint Graph Augmentation and Adaptive Synthetic Sampling for Imbalanced Node Classification NSTL国家科技图书文献中心

Guangquan Lu |  Wanxin Chen... -  《Natural Language Processing and Chinese Computing,Part IV》 -  CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 469~482 - 共14页

摘要: with node classification problems and has achieved |  the number of node classes is balanced. In fact, in |  the classification performance of GNNs. This paper |  proposes a new framework GraphAdasyn for imbalanced node |  classification to address this problem. The model inputs a
关键词: Graph representation learning |  Imbalanced node classification |  Graph neural network

8. CDCGAN: Class Distribution-aware Conditional GAN-based minority augmentation for imbalanced node classification NSTL国家科技图书文献中心

Liu, Bojia |  Zheng, Conghui... -  《Neural Networks》 - 2025,183 - Article 106933~Article 106933 - 共15页 - 被引量:1

摘要:Node classification is a fundamental task of | , thus leading to unappealing classification |  node representations. To address this issue, we | , we extract the node embeddings and class |  last, to ensure the distinguishability of node
关键词: Graph neural networks |  Generative adversarial networks |  Imbalanced node classification

9. Node classification in the heterophilic regime via diffusion-jump GNNs NSTL国家科技图书文献中心

Begga, Ahmed |  Escolano, Francisco... -  《Neural Networks》 - 2025,181 - Article 106830~Article 106830 - 共12页

摘要:In the ideal (homophilic) regime of vanilla |  GNNs, nodes belonging to the same community have the |  same label: most of the nodes are harmonic (their |  unknown labels result from averaging those of their |  neighbors given some labeled nodes). In other words
关键词: Graph neural networks |  Heterophily |  Homophily |  Node classification |  Diffusion |  Dirichlet problem |  High-order graph neural networks |  Structural filters

10. A Federated Parameter Aggregation Method for Node Classification Tasks with Different Graph Network Structures NSTL国家科技图书文献中心

Hao Song |  Jiacheng Yao... -  《Data Security and Privacy Protection,Part I》 -  International Conference on Data Security and Privacy Protection - 2025, - 225~243 - 共19页

摘要:Over the past few years, federated learning |  has become widely used in various classical machine |  learning fields because of its collaborative ability to |  train data from multiple sources without compromising |  privacy. However, in the area of graph neural networks
关键词: Federated learning |  Graph neural networks |  Node classification
检索条件Node classification

NSTL主题词

  • NSTL学科导航