RAPID: Analyses of Emotions Expressed in Social Media and Forums During the COVID-19 Pandemic

研究人员正在构建一个综合数据库，从社交媒体和网络论坛中收集、存储和分析与COVID-19相关的恐惧、焦虑、悲伤和愤怒的相关内容。先前的文献表明，压力会降低决策能力和决策质量。随着社交媒体和基于网络的论坛成为主要的交流形式，在线风险（如隐私、安全漏洞）可能有所增加。这个数据集将建立一个与位置和时间相关的情绪记录，这些情绪可能与虚拟威胁的脆弱性增加有关。这些数据将允许对风险进行分析，例如在网上分享更多个人信息、错误信息的发起和传播、宽松的安全偏好以及内部威胁。第二个目标是回答有关本次大流行期间所经历的负面情绪与区域差异和社会经济地位之间关系的基本研究问题。鉴于许多在线数据源不归档数据或提供可用于分析的档案，本研究是紧迫和及时的。这项研究通过分析和理解情绪如何与大流行期间的地方和区域社会和地理指标联系起来，为大流行期间的社区应对和决策提供信息，从而推动了科学的发展。研究小组将从2019年12月31日至2020年12月31日在10-15个社交媒体和网络论坛上收集COVID-19数据。数据收集将从中国当局首次治疗肺炎病例（后来被称为冠状病毒）开始。调查人员将追踪对COVID-19的反应一年，以评估公众对大流行的情绪反应。为了检查不同地区的健康和经济差异，调查人员将分析地理位置的职位，并将数据与人口普查局调查的变量相结合。人工智能和数据科学技术将用于处理和分析在此工作中收集的大量异构数据。根据这些分析，可以制定政策，以改善大流行期间的预防、安全、隐私和其他公众理解和政策。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 investigators are constructing a comprehensive database that collects, stores, and analyzes content related to fear, anxiety, sadness, and anger associated with COVID-19 from social media and web-based forums. Prior literature has demonstrated that stress can reduce decision making capacity and quality. Online risks (e.g., privacy, security vulnerabilities) may have increased as social media and web-based forums have become a predominant form of communication. This dataset will establish a location- and time-linked record of emotions that may be associated with increased vulnerability to virtual threats. The data will permit analyses of risks such as sharing of more personal information online, misinformation initiation and spread, relaxed security preferences, and insider threat. A second goal is to answer fundamental research questions about the linkages of negative emotions experienced during this pandemic with regional variation and socioeconomic status. This research is urgent and timely given that many online data sources do not archive data or make archives available for analyses. This research advances science by informing community response and policymaking during pandemics through an analysis and understanding of how emotions are linked to local and regional social and geographic indicators during the pandemic.The research team will collect COVID-19 data from 10-15 social media and web-based forums from December 31, 2019 to December 31, 2020. Data collection will begin when Chinese authorities first treated pneumonia cases that later became known as the coronavirus. The investigators will follow responses to COVID-19 for a year to assess the public's emotional responses to the pandemic. To examine health and economic disparities by region, the investigators will analyze geolocated posts and integrate the data with variables from the Census Bureau Survey. Artificial intelligence and data science techniques will be used in processing and analyzing the large amounts of heterogenous data collected in this effort. From these analyses, policies could be created to improve prevention, security, privacy and other public understanding and policies during the pandemic.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.