Berkeley Data Science Team Reflects on Tackling COVID Outbreaks

BIDS Faculty Affiliate Bin Yu is featured in a recent CDSS News article that recounts experiences early on in the pandemic, in the spring of 2020, when she was approached by a new nonprofit organization seeking data science expertise in its efforts to efficiently distribute personal protective equipment (PPE) to where it was needed most. Yu – professor of statistics and electrical engineering and computer sciences at UC Berkeley – quickly recruited students from her group, and in the end, the team’s predictions helped inform the shipment of at least 349,000 face shields to doctors and healthcare workers at a time when they were desperately needed. 

Yu and PhD Student Chandan Singh were recently invited to submit a paper to a special issue of Statistical Science on Data Science in a Time of Crisis: Lessons from the Pandemic. Their paper – Seven Principles for Rapid-Response Data Science: Lessons Learned from Covid-19 Forecasting – relates the team's experiences and distills seven principles they used to tap into domain knowledge (epidemiology) and medical logistics chains, curate a relevant data repository, develop models for short-term county-level death forecasting (in the US), and build a website for sharing visualization (an automated AI machine):

1. The decision to engage: preparedness and willingness.
2. Effective human organization: divide and conquer.
3. Gathering data and context: scraping, human contacts, and media reports. 
4. Data quality control: in-house data cleaning and curation. 
5. Speedy development and validation of many prediction algorithms. 
6. Uncertainty: measurement and empirical validation. 
7. Communicating results: interactive visualizations, open-source code, and a web interface. 

“Although these principles are described in the context of working with Response4Life, many of the principles overlap with those in standard data-science,” Yu said. “But in this project, we emphasized the need for a rapid response due to the fast spread of lethal COVID-19 hot spots across the country. There were no other county-level prediction models available in the U.S. until after our paper was submitted on May 16, 2020.”

Statistics-Computer Sciences Team Reflects on Tackling COVID Outbreaks
May 26, 2022  |  Jon Bashor  |  CDSS News