Research Publication Announcement: AI-Based Detection of Crypto-Asset Price Surges Using Transaction Network Features (DP No. 25-E-113)
- 有加 岡本
- Dec 25, 2025
- 1 min read
We are pleased to announce that the results of a collaborative research project conducted by Professor Yuichi Ikeda and his collaborators have been published as a Discussion Paper by the Research Institute of Economy, Trade and Industry (RIETI).
This study proposes an artificial intelligence (AI) framework for detecting price surges in crypto-asset markets by leveraging structural features of transaction networks. Using methods from complex network analysis and causal inference, the framework identifies network dynamics that precede price fluctuations and quantifies the probability of price surges. In addition, the study develops a method to retrospectively identify individual nodes that significantly contribute to price surges. The findings provide a novel analytical approach to risk management and regulatory oversight in crypto-asset markets.
Paper Information
DP Number: 25-E-113
Title:Artificial Intelligence for Detecting Price Surges Based on Network Features of Crypto Asset Transactions
Authors
Yuichi Ikeda (Kyoto University) / Hideaki Aoyama (Faculty Fellow) / Tetsuo Hatsuda (RIKEN) / Tomoyuki Shirai (Kyushu University) / Taro Hasui (Kyushu University) / Yoshimasa Hidaka (Kyoto University) / Krongtum Sankaewtong (Kyoto University) / Hiroshi Iyetomi (Rissho University) / Yuta Arai (Reitaku University) / Abhijit Chakraborty (Indian Institute of Science Education and Research) / Yasushi Nakayama (SBI Financial and Economic Research Institute) / Akihiro Fujiwara (Chiba Institute of Technology) / Pierluigi Cesana (Kyushu University) / Wataru Soma (Rissho University)
Related Links
Japanese Summary https://www.rieti.go.jp/jp/publications/summary/25120004.html
English Summary https://www.rieti.go.jp/en/publications/summary/25120004.html
Non-Technical Summary https://www.rieti.go.jp/jp/publications/nts/25e113.html
Project Page https://www.rieti.go.jp/jp/projects/program_2020/pg-05/014.html