Website: policychangeindex.org
Core maintainers:
- Julian TszKin Chan (julian.chan [AT] policychangeindex.org);
- Weifeng Zhong (weifeng.zhong [AT] policychangeindex.org).
External contributors:
- Kwan-Yuet "Stephen" Ho (KwanYuet.Ho [AT] leidos.com);
- Kit Lee;
- Kawai Leung.
Please email all comments/questions to weifeng.zhong [AT] policychangeindex.org
How severe was COVID-19 in China, really? It is widely suspected that the country's official numbers of diagnosed cases understate the extent of the outbreak, and even the official statistics themselves are incoherent. On February 13, 2020, the Chinese authorities confirmed more than 15,000 new cases nationwide---40 times the previous day's number---because of a change in counting criteria. Just two months after that, they revised the death toll for the city of Wuhan, the epicenter, upward by 50 percent, citing unspecified omissions.
The PCI-Outbreak uses a deep learning method to measure the severity of COVID-19 in China, not through the Chinese government's official numbers but through how state-controlled media talked about the outbreak. The algorithm is trained on SARS-episode articles in the People's Daily, the official newspaper of the Communist Party of China, to understand the tone and tenor of the SARS-episode narrative as the epidemic cycle evolved. The algorithm then assesses future outbreaks' severity against the SARS benchmark.
The PCI-Outbreak is built on the idea that words can speak louder than (some) numbers. While it may be simple to release false statistics, it is more difficult to conceal the truth when the government has to address a crisis, such as a severe infectious disease, at length in the media. Take the beginning of COVID-19 for example: When the Chinese government announced a lockdown of Wuhan, a city with a population of 11 million, and its neighboring cities and discussed the necessity of doing so in state media, the Chinese authorities had confirmed a total of fewer than 600 cases of the novel coronavirus across the country. Therefore, changes in words during the outbreak may provide us with a clearer picture of the severity than official numbers do.
For details about the methodology and findings of this project, please see the following research paper:
- Chan, Julian TszKin, Kwan-Yuet Ho, Kit Lee, Kawai Leung, and Weifeng Zhong. 2020. "Words Speak Louder Than Numbers: Estimating China's COVID-19 Severity with Deep Learning" Mercatus Working Paper (latest version available here).
Results will change as the underlying models improve. A fundamental reason for adopting open source methods in this project is so that people from all backgrounds can contribute to the models that our society uses to assess and predict changes in public policy; when community-contributed improvements are incorporated, the model will produce better results.
The first step for everyone (users and developers) is to open a free GitHub account. And then you can specify how you want to "watch" the PCI repository by clicking on the Watch button in the upper-right corner of the repository's main page.
The second step is to get familiar with the PCI-Outbreak repository by reading the documentation.
If you want to ask a question or report a bug, create a new issue here and post your question or tell us what you think is wrong with the repository.
If you want to request an enhancement, create a new issue here and provide details on what you think should be added to the repository.
Please cite the source of the latest Policy Change Index for Outbreak (PCI-Outbreak) by the website: https://policychangeindex.org.
For academic work, please cite the following research paper:
- Chan, Julian TszKin, Kwan-Yuet Ho, Kit Lee, Kawai Leung, and Weifeng Zhong. 2020. "Words Speak Louder Than Numbers: Estimating China's COVID-19 Severity with Deep Learning" Mercatus Working Paper (latest version available here).