Data and Code for "Comparing the Effects of Euclidean Distance Matching and Dynamic Time Warping in the Clustering of COVID-19 Evolution"

This repository contains the datasets and data sources, analysis code, and workflow associated with the manuscript "Comparing the Effects of Euclidean Distance Matching and Dynamic Time Warping in the Clustering of COVID-19 Evolution". The following resources are provided:

Data Files:

time_series_data.csv: A curated time series dataset with dates as rows and NUTS 2 regions as columns. Each column is labeled using a 4-letter abbreviation format "CC.RR", where "CC" represents the country code and "RR" represents the region code. This same abbreviation is also included in the accompanying GeoJSON file.

geometry_data.geojson: A GeoJSON file representing the spatial boundaries of the NUTS 2 regions, with the same 4-letter abbreviations used in the CSV file. EPSG:4326.

COVID19_data_sources.xlsx: This Excel file contains important metadata regarding the sources of COVID-19 data used in this study. It includes:

Source of the data for each country

Official website(s)

The agency responsible for the data

Description of the processing steps used to curate the data into the final time series.

Code:

analysis.py: A Python script used to process and analyze the data. This code can be run using Python 3.x. The libraries required to run this script are listed in the first lines of the code. The code is organized in different numbered sections (1), (2), ... and sub-sections (1a), (1b) ... Make sure to run the script one (sub-)section at a time, so that everything stays overviewable and you don't get all the output at once.

Workflow:

workflow.png : A detailed workflow according to the Knowledge Discovery in Databases (KDD) process, outlining the steps involved in processing and analyzing the data, including the methods used. This workflow provides a comprehensive guide to reproducing the analysis presented in the paper.