RAPID: Leveraging Twitter Data for Real-time Public Health Responses to Coronavirus: Identifying Affective Desensitization, Loneliness and Depression, and Trust

该项目的目标是利用Twitter来揭示与COVID-19传播相关的社会因素如何潜在地破坏公共卫生建议，同时也为公共卫生沟通提供有关预防的信息。这项研究将导致(1)发展与流行病的社会影响和有效传播有关的新理论，(2)理解信息内容如何导致对这种现象的信任和不信任；(3)描述大流行不同阶段出现的广泛的公民叙述；(4)可视化随着疾病在时间和空间上的传播，社会因素的动态是如何演变的。结果将迅速传达给相关的公共卫生机构，并将有助于使公共卫生宣传在公民中更加有效和可信。鉴于该疾病的迅速传播，必须立即确定社会影响，以便公共卫生机构和组织能够迅速、动态和更有效地适应和应对，以建立信任。推特数据提供了一种识别冠状病毒大流行的社会模式的手段，因为它提供了公民反应的信号，以及需要卫生专业人员指导的地方，所有这些都是实时的。本研究将使用关键字、机器学习、主题建模和定性分析，对涉及持续数据收集的与冠状病毒相关的两个大规模推文数据集进行分析。研究小组将增强基于地图的可视化系统，以显示不同地理区域的研究问题随时间的聚合和注释的社会反应。这将使研究人员能够深入了解这种社会现象明显的特定地区，并通过考虑与这些地区报告的冠状病毒病例相吻合的其他类型的信息，为分析提供信息。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The goal of this project is to use Twitter to uncover how social factors related to the spread of COVID-19 can potentially undermine public health recommendations but also inform public health communication about prevention. This research will lead to (1) the development of new theories related to social effects and effective communication of a pandemic, (2) understanding how message content leads to trust and distrust in the phenomenon; (3) characterizing widespread citizen narratives that emerge during different phases of the pandemic; and (4) visualizing how the dynamics of the social factors evolve as the disease spreads over time and space. The results will be swiftly communicated to relevant public health agencies, and will be valuable for making public health communications more effective and trustworthy among citizens. Given the rapid spread of the disease, it is critical to identify the social repercussions immediately so that public health agencies and organizations can adapt and respond quickly, dynamically, and more effectively to build trust. Twitter data provides a means to identify societal patterns of the coronavirus pandemic as it provides signals of citizen reactions and where there are needs for guidance from health professionals, all in real time. This research will use keyword, machine learning, topic modeling, and qualitative analysis on two large-scale datasets of tweets related to the coronavirus that involve ongoing data collection. The research team will enhance a map-based visualization system to show aggregated and annotated social responses from the research questions in different geographic regions over time. This will enable the researchers to gain insights about specific regions where such social phenomena are evident and can inform the analyses by considering other types of information that coincide with reported coronavirus cases in these regions.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.