Scuola Superiore Sant'Anna di Pisa- Institute of Economics, Piazza dei Martiri della Libertà, 3, 56127, Pisa, Italy.
IMT Alti Studi Lucca - Linkalab, Complex Networks Computational laboratory, Piazza San Francesco,19, 50500, Lucca, Italy.
Soc Sci Med. 2021 Jun;278:113940. doi: 10.1016/j.socscimed.2021.113940. Epub 2021 Apr 19.
In this paper,we present an original study on the use of social media data to analyze the structure of the global health networks (GHNs) relative to health organizations targeted to malaria, tuberculosis (TBC) and pneumonia as well as twitter popularity, evaluating the performance of their strategies in response to the arising health threats. We use a machine learning ensemble classifier and social network analysis to discover the Twitter users that represent organizations or groups active for each disease. We have found evidence that the GHN of TBC is the more mature, active and global. Meanwhile, the networks of malaria and pneumonia are found to be less connected and lacking global coverage. Our analysis validates the use of social media to analyze GHNs and to propose these networks as an important organizational tool in mobilizing the community versus global sustainable development goals.
在本文中，我们提出了一项使用社交媒体数据的原始研究，旨在分析与疟疾、结核病（TBC）和肺炎相关的卫生组织的全球卫生网络（GHN）的结构，以及推特的受欢迎程度，评估它们应对新出现的健康威胁的战略表现。我们使用机器学习集成分类器和社会网络分析来发现代表每个疾病的组织或团体的活跃推特用户。我们有证据表明，TBC 的 GHN 更为成熟、活跃和全球化。同时，发现疟疾和肺炎的网络连接性较差，缺乏全球覆盖。我们的分析验证了使用社交媒体来分析 GHN 的方法，并提出这些网络是动员社区实现全球可持续发展目标的重要组织工具。
