作者: Martin J. Westgate , David B. Lindenmayer
DOI: 10.1111/COBI.12890
关键词: Scientific writing 、 Semantic network 、 Systematic review 、 Meta-Analysis as Topic 、 Computer science 、 Data science 、 Selection (linguistics) 、 Resource (project management) 、 Scientific literature 、 Process (engineering)
摘要: The need for robust evidence to support conservation actions has driven the adoption of systematic approaches research synthesis in ecology. However, applying review complex or open questions remains challenging, and this task is becoming more difficult as quantity scientific literature increases. We drew on science linguistics guidance why process identifying sorting information during so labor intensive, provide potential solutions. Several linguistic properties peer-reviewed corpora—including nonrandom selection topics, small-world semantic networks, spatiotemporal variation word meaning—greatly increase effort needed complete process. Conversely, resolution these complexities a common motivation narrative reviews, but rarely enacted with rigor applied analysis. Therefore, provides unifying framework understanding some key challenges highlights 2 useful directions future research. First, cases where complexity generates barriers synthesis, ecologists should consider drawing existing methods—such natural language processing construction thesauri ontologies—that tools mapping resolving that complexity. These could help individual researchers classify material manner valuable topic. Second, perspective writing rich resource worthy detailed study, an observation can sometimes be lost search data meta-analysis. For example, networks reveal redundancy complementarity among concepts, leading new insights questions. Consequently, wider may facilitate improved richness synthesis.