| Project/Area Number |
19K20352
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 61030:Intelligent informatics-related
|
|---|
| Research Institution | National Institute of Advanced Industrial Science and Technology |
|---|
Principal Investigator |
LIU XIN 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, 主任研究員 (20803935)
|
|---|
| Project Period (FY) |
2019-04-01 – 2021-03-31
|
| Project Status |
Completed (Fiscal Year 2020)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2020: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2019: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
|
|---|
| Keywords | Graph embedding / Graph neural network / Bipartite graph / Heterogeneous graph / Knowledge graph / Node classification / Node ranking / Graph Embedding / Graph Neural Network / Node Classification / Node Ranking / Bipartite Graph / Knowledge Graph / Community Detection / Recommender System / Graph Neural Networks / Social Networks |
|---|
| Outline of Research at the Start |
Learning low-dimensional vector representations for graph nodes, or graph embedding, is a key step for graph analysis. The goal of this project is to develop a unified methodological framework for embedding the various graphs, which are proper representations of real-world complex systems. We will base our approach to recent advances of graph embedding and study the bipartite graphs by fully considering the distinctive features. The outcomes of this research will facilitate the graph analysis and provide insights into how to make good use of the information hidden in the various graphs.
|
| Outline of Final Research Achievements |
The outcomes of the project involve both the research methods as well as open-source softwares. We have proposed embedding approaches for different graphs, including the unipartite graphs, the bipartite graphs, the heterogeneous knowledge graphs, the attributed graphs, and the multigraphs. We also have successfully applied our approaches to real-world applications, including forecasting ambulance demand, citation recommendation, and brain disorder diagnosis.
|
| Academic Significance and Societal Importance of the Research Achievements |
このプロジェクトでは、様々なグラフで、グラフ埋め込みの共通の本質を明らかにします。これまでに、単部グラフ、二部グラフグラフ、異種頂点グラフ、ナレッジグラフ、属性付きグラフ、マルチグラフなど、さまざまなグラフに対する埋め込み手法を提案してきました。さまざまなグラフを研究することは、実世界の複雑なシステムを適切に表現し、幅広い応用につながるため、科学的に意義があります。
|
