COVID-19-DBs

Datasets

Name	Summary	Dataset	Num. Rows and Columns
DB1	Disease Related Data	confirmed_case.csv death.csv	<3208, 434> <3208, 434>
DB2	Demographic and Mobility Data	demographics.csv	<31140, 7>
DB3	Reddit Data	post.csv comment.csv	<72319, 12> <602862, 9>
DB4	Constructed Graph	graph.csv	<103243, 3>
DB5	Vaccination Data	vaccination.csv	<64, 116>

Detailed Descriptions of Datasets

Name	Description
DB1	Disease Related Data: We have collected the up-to-date county-based coronavirus related data (i.e., the number of confirmed cases and deaths) from official public health organizations and digital media with real-time updates of COVID-19. By the date, we have collected these data for 3,208 counties from 52 states (including Washington, D.C. and Puerto Rico) on a daily basis from February 28, 2020 to date (i.e., May 05, 2021).
DB2	Demographic and Mobility Data: We parse the demographic data collected from the United States Census Bureau in a hierarchical manner: for each city, county or state in the U.S., the dataset includes its estimated population, population density (e.g., number of people per square mile), age and gender distributions. By the date, we have made the demographic and mobility dataset available for public use, including the information of estimated population, population density, and GPS coordinates for 31,140 cities, 3,208 counties and 52 states (including Washington, D.C. and Puerto Rico).
DB3	Reddit Data: We initialize our efforts on social media data with the focus of public perception analysis on Reddit, as it provides the platform for scientific discussion of dynamic policies, announcements, symptoms and events of COVID-19. We have collected and analyzed the data from 48 state-based subreddits (i.e., Washington, D.C. and 47 states). By the date, we have crawled and automatically analyzed 72,319 posts by 20,284 users on Reddit associated with 602,862 comments by 68,688 users on the discussion of COVID-19 from February 17, 2020 to date (i.e., February 17, 2021). Along with these data, this publicized dataset also includes the sentiment (0: negative, 1: positive) of each post and comment.
DB4	Constructed Graph: Based on the designed network schema in our paper (see below reference), the constructed graph has 34,401 nodes (i.e., 1 node with type of nation, 52 nodes with type of state, 3,208 nodes with type of county, 31,140 nodes with type of city) and 103,243 edges.
DB5	Vaccination Data: We have collected the up-to-date vaccination data (i.e., total vaccinations per hundred) from the Centers for Disease Control and Prevention. By the date, we have collected the data for 64 states/territories/federal entities from January 12, 2021 to date (i.e., May 06, 2021).

Column Details

• confirmed_case.csv, death.csv

["county name", "2/28/2020",...,"5/5/2021"]

• demographics.xlsx

["city", "county_name", "state_name", "lat", "lng", "population", "density"]

• post.csv

["id", "title", "author", "content", "post_type", "created_time", "num_of_comments", "attachment", "post_url", "subreddit", "state", "sentiment (0: negative, 1: positive)"]

• comment.csv

["comment_id", "post_id", "author", "content", "created_time", "comment_url", "subreddit", "state", "sentiment (0: negative, 1: positive)"]

• graph.csv

["entity", "entity", "relation (0: include, 1: near)"]

• vaccination.csv

["State/Territory/Federal Entity", "1/12/2021",...,"5/6/2021"]

Reference

If you use our dataset(s), please cite our paper:

@article{Ye2020_Covid-19_Satellite,
  title={α-Satellite: An AI-driven System and Benchmark Datasets for Hierarchical Community-level Risk Assessment to Help Combat COVID-19},
  author={Ye, Yanfang and Hou, Shifu and Fan, Yujie and Qian, Yiyue and Zhang, Yiming and Sun, Shiyu and Peng, Qian and Laparo, Kenneth},
  journal={arXiv preprint arXiv:2003.12232},
  year={2020} 
}