RAPID: Near Real-Time Quantifiable Social Media Data for Improved Modeling, Tracking and Mitigating the Spread of the Ebola Virus

这项研究使开发新的定量危机地图能够近乎实时地估计当地埃博拉病毒传染性区域传播率，从而加强有效分配最近开发的有限可用疫苗，以减轻埃博拉的传播。它还将提供一种机制，通过与区域社会媒体观测结果进行比较，验证和改进疾病预防控制中心的业务预报模型。它可能允许在第三世界国家进行疾病监测和早期发现，这反过来将对全球产生影响，正如最近埃博拉疫情在全球迅速蔓延所证明的那样。这项工作应推广到其他传染性病毒条件。此外，这项研究将能够利用来自未来手持设备的社交媒体健康数据。更具体地说，该项目正在开发一种方法，利用和增强来自Twitter的可量化社交媒体数据，将其转化为流行病学感染病毒传播模型，目的是验证模型预测，并提供传统来源无法获得的埃博拉病毒病疑似病例的额外渠道。这项工作的基础是需要收集与三个撒哈拉以南非洲国家埃博拉暴露区域人口有关的实时定量媒体数据。虽然互联网站点上存在可公开获取的部分数据，但只有从管理和控制Twitter和Instagram等网站的组织购买才能获得完整的公共数据来源。所开发的方法将有助于评估基于社交媒体数据的模型预测的有效性，以便今后改进传染病传播和预防的预测。这项工作与FAU、FIU和UMN的研究人员合作，与各自的RAPID提案协调并并行提交，旨在共同解决具有国家和全球意义的公共卫生问题，特别是缓解突发公共卫生事件的大数据问题。具体而言，提案人将与FAU和FIU合作，通过使用创新的大数据分析技术，整合来自Twitter的可量化社交媒体数据，对埃博拉病毒的传播进行建模。除了目前的埃博拉问题之外，拟议的方法和建立联盟的努力旨在为广泛的公共卫生问题提供支助解决办法。

The research enables the development of new quantitative crisis maps to approximate local regional rates of infectious Ebola virus spreading in near real time thereby enhancing the effective distribution of limited availability of recently developed vaccines to mitigate the spread of Ebola. It will also provide a mechanism for validating and improving operational forecast models of the CDC by comparisons with regional social media observations. It may allow for disease surveillance and early detection among third-world countries, which will in turn have impacts globally as evidenced by the rapid global spread of the recent Ebola outbreak. The work should be generalizable to other infectious virus conditions. Furthermore the research will be capable of exploiting social media health data from future hand held devices.More specifically, this project is developing methodology for utilization and augmentation of quantifiable social media data from Twitter into epidemiological infectious virus spread model, with the aim to validate the model forecasts and provide additional channels of suspected EVD cases not available through traditional sources. The work is based on a need to collect real-time quantitative media data related to Ebola-exposed regional populations in the three sub-Saharan African countries. Although publicly available partial data exists on Internet sites, the complete source of public data is available only by purchase from organizations that manage and control such sites as Twitter and Instagram. The developed methods would contribute in evaluating the effectiveness of model forecast based on the social media data for future improvements in the prediction of infectious disease spread and prevention.This work, coordinated with and submitted in parallel with respective RAPID proposals, in coalition with researchers at FAU, FIU, and UMN, aims to jointly address public health problems of national and global significance, particularly the problems of big data for mitigation of public health emergencies. Specifically, the proposers will work with FAU and FIU on modeling Ebola spread through use of innovative big data analytics techniques, integrating quantifiable social media data from Twitter. Beyond the present issue of Ebola, the proposed methodology and the coalition-building effort aims to enable support solutions in a wide range of public health issues.