일시: 2025년 11월 4일 (화) 오후4시

장소: 경영대학1호관 308호 세미나실

연자: 정호현 교수님 (성신여자대학교 수리통계데이터사이언스학부)

강연제목: Measuring Rich-Get-Richer in Dynamic Networks: A Latent Space Approach

초록:

This study proposes a dynamic latent space network model to measure the rich-get-richer phenomenon while accounting for fit-get-richer effects. By introducing a popularity influence parameter in addition to latent node positions, the model separates intrinsic node fitness from cumulative advantage mechanisms in network growth. Inference is conducted within a Bayesian framework and validated through simulations. The model is then applied to Bitcoin trust networks (OTC and Alpha), revealing clear evidence of the rich-get-richer effect and allowing comparisons of its magnitude across platforms. Results show that although conventional model selection criteria (e.g., DIC) have difficulty identifying the true latent dimension, the measurement of the rich-get-richer effect remains robust across different latent space dimensions.