Inspiration: we feel human factors could be important indicators for the covid-19 spread and the analysis could lead to insightful discovery 

Approach: we look for data, analyze and clean the data, and then build visualization and modelling

Tools/software: we used Python, R, Tableau for data visualization, and built machine learning/deep learning-based prediction models such as Ridge regression and XGB.

Challenges: 

1. Communication in the virtual environment created a lot of challenges for teamwork

2. Discussion within limited time in a brand-new team made it difficult to reach team consensus

3. The vast data resources made it hard to find useful data

Achievements: 

1. We built a data model prototype that streamlines data analytics, integration and visualization.

2. We identified major policy factors that have a significant impact on coronavirus spread. For example, gathering ban is correlated to the Covid-19 trend.

3. We found the higher urbanization, the more coronavirus cases. 

4. We discovered gathering ban has the best effect among all policies.

5. We noticed that the unemployment population has a positive relationship with confirmed Covid-19 cases.

Future Work:

This working prototype can be further enhanced to provide more capable analytics/prediction assistance for policy making in fighting Covid-19.

https://prezi.com/view/v5pWfnY76PFnAKijpAYz/

Reference Data Source: https://earthdata.nasa.gov/

Policy: https://coronavirus-disasterresponse.hub.arcgis.com/

Population: https://www.census.gov/acs/www/data/data-tables-and-tools/ranking-tables/

Confirmed cases: https://github.com/CSSEGISandData/COVID-19

Policy: https://docs.google.com/spreadsheets/d/e/2PACX-1vRlJWZJ7OkGUW57_rdA2n3xBJ3qjW6u4Z9N6K9Y5L4bM_6H7-S308qdKmJfpVstYWf300nyujvZPFSy/pubhtml?urp=gmail_link#

Urbanization: https://www.researchgate.net/publication/229073021_NCHS_urban-rural_classification_scheme_for_counties

Urbanization classification rule: https://wonder.cdc.gov/wonder/help/CMF/Urbanization-Methodology.html