This study examined the geodiversity of research through comparing topic focus with author location using SDG 2: Zero hunger as a case study. As the research was related to hunger, papers were mapped on to the Global Hunger Index country categories as convenient classification. The publication dataset comprised 60,000 papers from the Dimensions database that have been associated with hunger research using Digital Science machine learning algorithm that enhances expert led search strategies. Only 41% hunger-related publications that focus on countries most affected by hunger feature authors affiliated to institutions in those countries. Even fewer of those publications feature locally based authors in first or last position. These numbers gradually reverse as the level of hunger declines. We analyse sample papers in an attempt to understand the reasons for these trends. These included differences in research infrastructure, sub-authorship recognition such as acknowledgements, and limitations of the relationship between country mention and real topical focus. We did not find evidence of widespread differences between senior and overall authorship and consequently urge caution before judging international collaborations as helicopter research based only on author country affiliations and authorship position.

翻译：本研究以可持续发展目标2“零饥饿”为案例，通过比较研究主题关注点与作者所在地，探讨了研究的区域多样性。由于研究涉及饥饿问题，我们将论文映射至全球饥饿指数国家分类作为便捷的分类标准。出版数据集包含来自Dimensions数据库的60,000篇论文，这些论文通过Digital Science机器学习算法与饥饿研究相关联，该算法增强了专家主导的搜索策略。在关注受饥饿影响最严重国家的研究成果中，仅有41%的论文作者隶属于这些国家的机构。这些论文中以当地作者担任第一作者或最后作者的比例更低。随着饥饿水平的下降，这些数值逐渐逆转。我们通过分析样本论文，试图理解这些趋势背后的原因，包括研究基础设施的差异、次署名认可（如致谢）问题，以及国家提及与实际研究主题焦点之间关系的局限性。我们没有发现高级作者与总体作者之间存在普遍差异的证据，因此敦促在仅依据作者国籍隶属关系和作者排序将国际协作判定为直升机研究时需谨慎。