| 研究者所属（当時） | 資格 | 氏名 | |
|---|---|---|---|
| （代表者） | グローバル・エデュケーション・センター | 講師 | 邵 騰飛 |
| （連携研究者） | 早稲田大学 | 教授 | Goto Masayuki |
| （連携研究者） | 早稲田大学 | 准教授 | Yuya Ieiri |
| （連携研究者） | 群馬大学 | 准教授 | Xu Wang |
- 研究成果概要
This research successfully developed a hypergraph motif analysis method to decode complex, non-linear structures in consumer evaluations. Unlike traditional pairwise analysis, our approach captures high-order interactions among consumers, products, and sentiments.
The project yielded three core contributions:
Methodological Innovation: Established a robust framework utilizing network motifs to model time-series dynamics and multi-dimensional consumer feedback.
Empirical Validation: The method was successfully applied to diverse real-world datasets, validating its effectiveness across second-hand luxury e-commerce, commercial district marketing (stamp rally data), and AI ethics education.
Market Insight Generation: The analysis revealed critical structural patterns, including dynamic sentiment dependencies and distinct behavioral differences between e-commerce and brick-and-mortar luxury markets.
These findings provide granular, actionable insights for data-driven marketing strategies. The project outcomes have been widely disseminated, resulting in three peer-reviewed journal articles (including Decision Analytics Journal and Journal of Information Processing) and one international conference proceeding in 2025.