作者: Martin J. Westgate , Philip S. Barton , Jennifer C. Pierson , David B. Lindenmayer
DOI: 10.1111/COBI.12605
关键词: Statistical hypothesis testing 、 Ecology 、 Biology 、 Topic model 、 Data science 、 Task (project management) 、 Latent Dirichlet allocation 、 Identification (information) 、 Scientific literature 、 Popularity 、 Systematic review
摘要: Keeping track of conceptual and methodological developments is a critical skill for research scientists, but this task increasingly difficult due to the high rate academic publication. As crisis discipline, conservation science particularly in need tools that facilitate rapid yet insightful synthesis. We show how common text-mining method (latent Dirichlet allocation, or topic modeling) statistical tests familiar ecologists (cluster analysis, regression, network analysis) can be used investigate trends identify potential gaps scientific literature. tested these methods on literature ecological surrogates indicators. Analysis popularity within corpus showed strong emphasis monitoring management fragmented ecosystems, while analysis suggested greater role genetic Our results automated text with care, provide information complementary given by systematic reviews meta-analyses, increasing scientists' capacity