This study examines how changes in research topics affect the academic productivity of a research group.
We selected researchers from the Mathematics Genealogy Project (MGP), a dataset comprising more than 250,000 researchers. Publication, citation, and H-index data were gathered from multiple sources — reliable measures of research performance and productivity for both individual researchers and research groups.
Researchers were organized into groups, and a topic model was used to investigate how the distribution of research topics within these groups evolved over time. We then evaluated the impact of frequent topic changes on academic output by analyzing the correlation between:
- Publication count
- Citation count
- H-index
- Variance in topic distribution (topic deviation)
- Across research groups as a whole, there is a weak negative correlation between topic deviation and academic output (citations, H-index, and publications).
- When research groups are segmented by their topic deviation values, a clearer pattern emerges: groups that maintain a consistent focus on a specific set of topics show better academic output — in terms of publications, citations, and co-author collaboration — compared to groups that frequently shift topics.
| Component | Description |
|---|---|
| Source dataset | Mathematics Genealogy Project (MGP), 250K+ researchers |
| Metrics collected | Publications, citations, H-index |
| Grouping | Researchers organized into academic groups |
| Modeling | Topic model applied to track topic distribution evolution over time |
| Analysis | Correlation analysis between topic deviation and academic output metrics |
If you use this work, please cite the corresponding paper (details to be added).