Graph Learning on Imperfect Data: Theoretical Foundations and Innovative Real-world Applications

• Project/Area Number: 25K03231
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 62020:Web informatics and service informatics-related; Basic Section 60080:Database-related; Sections That Are Subject to Joint Review: Basic Section60080:Database-related , Basic Section62020:Web informatics and service informatics-related
• 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)
• Co-Investigator(Kenkyū-buntansha): 村田 剛志 東京科学大学, 情報理工学院, 教授 (90242289)
• Project Period (FY): 2025-04-01 – 2029-03-31
• Project Status: Granted (Fiscal Year 2025)
• Budget Amount: ¥17,160,000 (Direct Cost: ¥13,200,000、Indirect Cost: ¥3,960,000); Fiscal Year 2028: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000); Fiscal Year 2027: ¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000); Fiscal Year 2026: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000); Fiscal Year 2025: ¥5,460,000 (Direct Cost: ¥4,200,000、Indirect Cost: ¥1,260,000)
• Keywords: Graph Machine Learning / Graph Neural Network / Graph Embedding / Graph Analysis / Imperfect Data

Outline of Research at the Start

Deep learning has achieved groundbreaking success in fields such as image recognition and natural language processing, significantly impacting industry. However, these techniques have not seen comparable success with graph data. A major reason is that current graph learning technologies typically require idealized data, which are rarely available and make real-world applications impractical. This research aims to develop approaches that maintain high performance on imperfect graph data, enhancing the practicality of graph learning and enabling innovative applications across diverse fields.