RAPID: #COVID-19: Understanding Community Response in the Emergence and Spread of Novel Coronavirus through Health Risk Communications in Socio-Technical Systems

弱势群体在新冠肺炎疫情发生和传播过程中的风险认知和风险规避行为是个体或群体与其社区内外网络社会邻居互动的时空功能，需要通过多种信息渠道（如广播、电视、互联网等传统渠道和/或社交媒体等非传统渠道）捕捉这种互动。快速反应研究（Rapid）项目的主要目标是利用数据驱动方法和网络科学理论，收集对时间敏感的在线社交媒体和众包数据，分析新型冠状病毒出现和传播过程中的健康风险沟通和社区反应模式。主要重点将是了解个人如何在网上受到社会影响，同时在疾病继续传播的过程中，在各自的社区中传达风险并进行互动。影响的概念将通过量化对这种传播行为的网络效应和描述在这种疾病爆发中与社会联系在一起并面临健康风险的人们之间如何交换信息来获得。鉴于社区对COVID-19的应对准备有限或没有准备，而且已经受到影响的社区的恢复时间不确定，而新的社区受到威胁，因此数据收集工作需要快速响应，以更好地覆盖和仔细监测。这些数据将包括参与COVID-19应对、恢复和缓解工作的受影响社区人员和政府官员的大规模短暂在线互动，然后是数据驱动的网络分析和所采取的COVID-19风险沟通策略和风险规避行为的信息图表。本研究不仅将扩展重大疾病爆发前后风险感知和传播策略的时空动态知识库，而且还将产生数据驱动的推理技术，以提高我们对人们如何表达各种担忧以及如何利用和嵌入这些信息来设计干预措施的理解。这项快速反应研究的方法和结果将有利于应急管理和公共卫生机构根据人们对COVID-19的反应以及他们在应对类似公共卫生危害时的社会网络特征、活动和互动，为有不同需求的公众制定有针对性的信息传播政策。公众参与风险沟通可以导致更有效的决策，并加强公众对监管过程的反馈。RAPID项目的主要目标是利用数据驱动方法和网络科学理论，挖掘和分析大规模时效性易逝的众包数据和社交媒体数据（丰富的时空数据），揭示新型冠状病毒发生和传播过程中的健康风险传播和社区应对模式。具体目标有三个：(1)记录在重大疾病爆发期间公众如何通过其在线社交网络互动和交流健康风险信息；(2)对多源数据进行认证，发现异常，避免信息过载和错误信息传播；(3)研究在线社交网络如何影响保护行动（例如，社交距离、自我隔离决定），即健康风险沟通中的信息级联。为实现目标和宗旨，该项目将利用用户、机构、新闻来源的短暂时间和地理标记社交媒体互动，并辅以有关COVID-19的众包信息。本研究将对文献有五个理论和方法上的贡献。它将：(1)促进我们对个人如何在网络上受到社会影响的理解，同时在疾病继续传播的过程中，在各自的社区中传达健康风险并进行互动；(2)通报在此类疾病暴发中，在线社交和面临健康风险的人群之间如何交换信息的文献；(3)使用新的机器学习和网络科学模型来量化对这种传播行为的影响和网络效应；(4)基于风险与疾病传染的时空相关性，捕捉网络构成、风险沟通策略和风险规避行为的变异性；(5)确保从多个来源收集的数据的真实性，并针对社会技术系统中的健康风险异常检测开发更准确的全分布式计算算法。这项研究的结果将有助于公共卫生和应急管理机构根据不同用户群体的社会网络特征、活动和互动来制定有效的信息传播政策，以应对类似的公共卫生危害。这项研究的方法和影响可用于设计有效的干预政策，适用于公共卫生风险成为主要关切的其他自然和人为灾害情况。该项目将通过优先考虑美国弱势群体和代表性不足的社区，为K-12、本科生和研究生提供创新的积极学习环境。该项目将训练学生使用高性能计算资源收集、存储、处理、分析和建模大规模数据所需的计算技能。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。

Risk perception and risk averting behaviors of vulnerable communities in the emergence and spread of COVID-19 are spatio-temporal functions of individual or group interactions with their online social neighbors within or outside their communities and such interactions need to be captured through diverse information channels (e.g. traditional outlets such as radio, television, internet and/or non-traditional outlets such as social media). The primary goal of this Rapid Response Research (RAPID) project is to collect time-sensitive online social media and crowd-sourced data and analyze patterns of health-risk communication and community response in the emergence and spread of novel Coronavirus using data-driven methods and network science theories. The major focus will be towards understanding how individuals are socially influenced online, while communicating risk and interacting in their respective communities as the disease continues to spread. The notion of influence will be captured by quantifying the network effects on such communication behavior and characterizing how information is exchanged among people who are socially connected online and exposed to health risk in such outbreaks of disease. Given that communities responded to COVID-19 with limited or no preparation and there is uncertainty in the length of recovery for the communities already affected while new communities being threatened, the data collection effort requires rapid response for better coverage and careful monitoring. The data will include large-scale ephemeral online interactions of people in the affected communities and public officials who are involved in COVID-19 response, recovery, and mitigation efforts, followed by a data-driven network analytics and infographics of COVID-19 risk communication strategies and risk averting behaviors adopted. The proposed research will not only expand the knowledge base of spatio-temporal dynamics of risk perception and dissemination strategies in the emergence and aftermath of a major disease outbreak, but will also result in data-driven inference techniques to improve our understanding of how people express diverse concerns and how to harness and embed such information for designing intervention measures. The methodologies and findings of this rapid response research will benefit emergency management and public health agencies to define targeted information dissemination policies for public with diverse needs based on how people reacted to COVID-19 and their social network characteristics, activities, and interactions in response to similar public health hazards.Public engagement in risk communication can lead to more effective decision-making and enhanced public feedback to the regulatory process. The primary goal of this RAPID project is to mine and analyze large-scale time-sensitive perishable crowd-sourced and social media data (rich spatio-temporal data) and reveal patterns of health-risk communication and community response in the emergence and spread of novel Coronavirus using data-driven methods and network science theories. The specific aims are threefold: (1) to document how public interact and communicate health risk information through their online social networks during a major disease outbreak; (2) to authenticate data from multiple sources and detect anomalies to avoid information overload and spread of misinformation; and (3) to examine how online social networks influence protective actions (e.g., social distancing, self-quarantine decisions) i.e. information cascades in health risk communication. To achieve the goal and aims, the project will utilize ephemeral time and geo-tagged social media interactions of users, agencies, news sources supplemented with crowd-sourced information on COVID-19. This study will have five theoretical and methodological contributions to the literature. It will: (1) advance our understanding of how individuals are socially influenced online, while communicating health risks and interacting in their respective communities as the disease continues to spread; (2) inform the literature on how information is exchanged among people who are socially connected online and exposed to health risk in such outbreaks of disease; (3) use novel machine-learning and network science models to quantify influence and network effects on such communication behavior; (4) capture the variability in network composition, risk communication strategies and risk averting behaviors adopted based on spatio-temporal correlations of risk and disease contagion; (5) ensure authenticity of the collected data from multiple sources and develop more accurate fully-distributed computational algorithms tailored to health risk anomaly detection in socio-technical systems. The findings from this research will be useful to public health and emergency management agencies for tailoring effective information dissemination policies for diverse user groups based on their social network characteristics, activities, and interactions in response to similar public health hazards. The methodologies, and implications of this research can be transferred in designing effective intervention policies to other natural and man-made disaster contexts in which public health risks become major concerns. The project will engage, mentor, and offer an innovative active learning environment for K-12, undergraduate, and graduate students by giving priority to disadvantaged and underrepresented communities in USA. The project will train students on computational skills required for collecting, storing, processing, analyzing and modeling large-scale data using high performance computational resources.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.