Published in arXiv (Working Paper/Preprint), 2026
Proposes a statistical framework combining topic modelling with trend analysis to detect and characterise the emergence of narratives in longitudinal text corpora.
Recommended citation: Medeiros, C., Quigley, J., & Revie, M. (2026). A statistical framework for detecting emergent narratives in longitudinal text corpora. arXiv:2602.20939.
Download Paper
Published in Bayesian Young Statisticians Meeting (BaYSM 2025), Chiba, Japan, 2025
Monte Carlo simulation study examining how corpus characteristics affect LDA estimation performance, and comparing LDA and STM under temporally evolving topic prevalence. Presented as a poster at BaYSM 2025.
Recommended citation: Medeiros, C., Quigley, J. & Revie, M. (2025). A simulation study of topic model performance under varying corpus conditions. Poster presented at BaYSM 2025, Chiba, Japan.
Published in Working Paper, 2024
Applies Structural Topic Models to compare academic entrepreneurial ecosystem discourse with policy documentation across the four UK nations, examining the extent to which research findings are reflected in policy language.
Recommended citation: Medeiros, C. & Green, S. (2024). Are we speaking the same language? Using topic models to shed light on research utilization issues between entrepreneurial ecosystem scholars and policy-makers. Working paper.