RAPID: Tackling the Psychological Impact of the COVID-19 Crisis

根据正在进行的COVID-19大流行期间的要求，就地避难所的物理隔离强调了心理健康。它促使人们通过社交媒体联系起来。虽然社交媒体平台可以实现在线联系，但它们可以耸人听闻地讲述一些故事，忽略其他故事，煽动焦虑和恐惧。该项目将使用人工智能来分析社交媒体数据，并对心理健康、痛苦和脆弱性进行建模。它将提供工具，帮助了解社区社交焦虑与附近COVID-19疫情的关系。这项工作的成果有可能支持公共卫生组织(1)及时和积极地应对受COVID-19危机影响的社区的心理需求和需求；(2)集思广益，通过资源分配和优先排序，应对与COVID-19相关的焦虑经历，改善人们的生活质量。该项目将评估COVID-19大流行的影响，并通过以下方式提高国家的复原力：(1)开发数据和理论驱动的科学计算方法，以识别基于社交媒体的语言和社会网络标记，这些标记与美国境内受影响社区的COVID-19危机相关的焦虑、压力和其他心理健康衰退有关；(2)建立预测模型，预测哪些社区最容易受到这些心理衰退的影响；(3)利用疾病传播的流行病学模型，全面了解社区的在线活动及其与病毒接近程度的离线时空地理背景；(4)开展以人为本的评价；(5)提供数据、开源工具包和数据驱动的演示，说明特定社区如何容易受到COVID-19大流行的影响，以支持公共卫生工作者和公众制定及时和积极的干预措施。总的来说，通过跨学科领域专家的反复参与，新的计算工件将改变COVID-19应对措施，同时考虑到危机的更大社会技术背景。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 physical isolation of shelter-in-place, as demanded during the ongoing COVID-19 pandemic, stresses psychological well-being. It pushes people to connect via social media. While social media platforms enable online connection, they can sensationalize some narratives and ignore others, fomenting anxiety and fear. This project will use artificial intelligence to analyze social media data and model psychological wellbeing, distress, and vulnerability. It will provide tools to help understand community social anxiety in relationship to nearby COVID-19 outbreaks. The outcomes of this work have the potential to support public health organizations in (1) responding to the psychological needs and demands of communities affected by the COVID-19 crisis in a timely and proactive fashion; and (2) brainstorming strategies to counter experiences of COVID-19 related anxiety and improve people’s quality of life through resource allocation and prioritization.This project will assess COVID-19 pandemic impacts and improve the nation’s resilience by: (1) developing data- and theoretically-driven scientific computational methods to identify social media based linguistic and social network markers associated with COVID-19 crisis-related anxiety, stress, and other downturns in psychological wellbeing in affected communities within the United States; (2) developing predictive models to forecast which communities will be most vulnerable to these psychological downturns; (3) leveraging epidemiological models of disease spread to derive holistic views of communities’ online activity and their offline spatiotemporal geographical context in relationship to their proximity to the virus; (4) conducting a human-centered evaluation; and (5) providing data, an open-source toolkit, and data-driven presentations of how particular communities are vulnerable to the COVID-19 pandemic to support public health workers and the general public in creating timely and proactive interventions. On the whole, through the iterative involvement of transdisciplinary domain experts, new computational artifacts will transform the COVID-19 response, taking into account the larger sociotechnical context of the crisis.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.