A decade of vertebrate palaeontology research: global taxa distribution, gender dynamics and evolving methodologies
Saved in:
| Title: | A decade of vertebrate palaeontology research: global taxa distribution, gender dynamics and evolving methodologies |
|---|---|
| Authors: | Wang, Haohan, Sterli, Juliana, Dupret, Vincent, Blom, Henning, Dr, 1969, Berta, Annalisa, Turner, Susan, Han, Daoming, Xu, Luyan, Pan, Zhaohui |
| Source: | Royal Society Open Science. 12(5) |
| Subject Terms: | vertebrate palaeontology, bibliometric analysis, gender gap, methodological evolution, latent Dirichlet allocation topic modelling |
| Description: | Using 12 104 publications from 2014 to 2023 in the DeepBone database, this study employs bibliometric methods, including full-text latent Dirichlet allocation (LDA) modelling, co-occurrence network analysis and geographic mapping with ArcGIS, to examine three key aspects of vertebrate palaeontology development: geographic distribution of newly established taxa, gender demographics among researchers and research trends. Gender data were analysed using automated tools with manual verification to ensure accuracy, while methodological evolution was investigated through systematic text mining and classification. Among 8336 newly established taxa, mammals (34.72%) and fishes (29.76%) dominate, followed by reptiles (25.34%), birds (7.39%) and amphibians (2.80%). Geographic analysis reveals significant regional disparities, with the USA (13.50%) and China (13.32%) contributing the most, while Africa and Oceania remain under-represented (less than 10%). Gender analysis indicates a gradual increase in female representation from 22.78 to 27.20% over the decade, highlighting the imperative to address gender disparities in vertebrate palaeontology, thereby advancing equity in alignment with UNESCO Sustainable Development Goal 5. LDA topic modelling identifies 15 distinct research topics, encompassing evolutionary biology, cranial and skeletal morphology, dinosaur-bird evolution and human evolution, while co-occurrence analysis highlights the evolution of research methodologies, revealing strong interconnections between phylogenetic analysis (15%), traditional morphological analysis (12%) and high-resolution imaging techniques (9%). |
| File Description: | electronic |
| Access URL: | https://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-557154 https://doi.org/10.1098/rsos.250263 |
| Database: | SwePub |
Full Text 
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstract: | Using 12 104 publications from 2014 to 2023 in the DeepBone database, this study employs bibliometric methods, including full-text latent Dirichlet allocation (LDA) modelling, co-occurrence network analysis and geographic mapping with ArcGIS, to examine three key aspects of vertebrate palaeontology development: geographic distribution of newly established taxa, gender demographics among researchers and research trends. Gender data were analysed using automated tools with manual verification to ensure accuracy, while methodological evolution was investigated through systematic text mining and classification. Among 8336 newly established taxa, mammals (34.72%) and fishes (29.76%) dominate, followed by reptiles (25.34%), birds (7.39%) and amphibians (2.80%). Geographic analysis reveals significant regional disparities, with the USA (13.50%) and China (13.32%) contributing the most, while Africa and Oceania remain under-represented (less than 10%). Gender analysis indicates a gradual increase in female representation from 22.78 to 27.20% over the decade, highlighting the imperative to address gender disparities in vertebrate palaeontology, thereby advancing equity in alignment with UNESCO Sustainable Development Goal 5. LDA topic modelling identifies 15 distinct research topics, encompassing evolutionary biology, cranial and skeletal morphology, dinosaur-bird evolution and human evolution, while co-occurrence analysis highlights the evolution of research methodologies, revealing strong interconnections between phylogenetic analysis (15%), traditional morphological analysis (12%) and high-resolution imaging techniques (9%). |
|---|---|
| ISSN: | 20545703 |
| DOI: | 10.1098/rsos.250263 |