COVID-19 | University of Tübingen

Enhanced COVID-19 data for improved prediction of survival

Supplementary files

This page contains data and code used in our paper: Wenhuan Zeng, Anupam Gautam and Daniel H. Huson (2021) On the Application of Advanced Machine Learning Methods to Analyze Enhanced, Multimodal Data from Persons Infected with COVID-19, Computation 9(4)

.

Data:

Initial COVID-19 dataset 183 Cases:	Initial.xlsx
Country-wise base and weighted average polarity score:	Polarity.csv
Past 14 day weather for each subject:	WeatherInfoTotal.csv
Weather, numerical values:	WeatherEmbeddingfile.xlsx
Enhanced COVID-19 dataset:	EnhancedCOVID_19Dataset.csv
Data downloaded from WHO website:	CSV_as_at_09_April_2020-Full_database.xlsx
Processed WHO publication data:	WHOJournalFormatted.txt
Country-wise mortality:	detail.xlsx

Code:

Python script for parsing WHO database:	PublicationListProcessing.py
Usage:	PublicationListProcessing.py -i InputFileName -o OutputFileName -t NumberOfThreads
Input file:	CSV_as_at_09_April_2020-Full_database.xlsx (Downloaded from WHO database)
Output file:	Text file with DOI and Institute Names
Required Python modules:	BeautifulSoup, pandas, requests, multiprocessing, argparse

Python script for crawling climate information on weather underground weather:	CrawlWeather.py
Usage:	CrawlWeather.py --input InputFileName --browser BrowserDriverName --output OutputFileName
Input file:	Initial.xlsx
Output file:	CSV file with Sample ID and Climate Information, each sample with 15 records
Required Python modules:	BeautifulSoup, pandas, NumPy, selenium, argparse