Modelling trend life cycles in scientific research using the Logistic and Gompertz equations

Abstract Scientific topics vary in popularity over time. In this paper, we model the life cycles of 200 trending by fitting Logistic and Gompertz models to their frequency time published abstracts. Unlike other work, use are algorithmically extracted from large datasets abstracts covering computer science, particle physics, cancer research, mental health. We find that produces lower median error, leading us conclude it is more appropriate model. Since asymmetric, with a steep rise followed long tail, implies scientific follow similar trajectory. also explore case double-peaking curves some cases, will peak multiple times as interest resurges. Finally, when looking at different disciplines, lifespan longer disciplines (e.g. research health) than others, which may indicate differences process culture between these disciplines.