2023 Fiscal Year Annual Research Report
Neural Network based Graph Learning: Model Evolution and Real-World Application
| Project/Area Number |
21K12042
|
|---|
| Research Institution | National Institute of Advanced Industrial Science and Technology |
|---|
Principal Investigator |
劉 欣 国立研究開発法人産業技術総合研究所, 情報・人間工学領域, 上級主任研究員 (20803935)
|
|---|
| Project Period (FY) |
2021-04-01 – 2024-03-31
|
| Keywords | Graph Neural Network / Graph Embedding / Graph Machine Learning / Dynamic Graph |
|---|
| Outline of Annual Research Achievements |
We have achieved four main outcomes. a) We proposed a new GNN model for robust predictions irrespective of the presence or absence of noise in edges, nodes, or both. This work has been accepted by PAKDD-2024. b) We proposed a continuous-time dynamic graph learning model to detect potential real-time donations in YouTube live streaming services. This work has been published in Machine Learning journal. c) We introduced an advanced multi-layer temporal graph neural network framework to learn entity representations and predict trends in social media. This work has been published in Complex & Intelligent Systems journal. d) We presented a knowledge graph embedding-based method for automatically predicting missing human biography records in Wikipedia. This work has been published in ECAI-2024.
|
