Here is a link for dataset and the paper can be found here.
The dataset is collected from Twitter using the scripts present in the "data_collection_scripts.zip" folder. You can read the detail about data collection pipeline in the section 2 of paper.
| Country | Data |
|---|---|
| USA | Yes |
| Spain | Yes |
| Italy | Yes |
| France | Yes |
| Germany | Yes |
| UK | Yes |
| Turkey | Yes |
| Russia | Yes |
| Brazil | Yes |
| Belgium | Yes |
| Canada | Yes |
| Netherlands | Yes |
| Switzerland | Yes |
| Portugal | Yes |
| India | Yes |
| Peru | Yes |
| Ireland | Yes |
| Austria | Yes |
| Sweden | Yes |
| Israel | Yes |
| Iran | Unavailable |
| China | No |
- The data cleaning scripts are present with most of the countries' data in their respective folder
- The scripts are in basic Python and JavaScript, and the rudimentary knowledge of programming is sufficient to understand and modify them according to need.
The scripts which are used for analysis, such as word-cloud, sentiment analysis, topic modelling(detail present in the section 3 of the paper) etc, can be found in code folder.
- The basic domain or programming knowledge is sufficient to understand the scripting for the project.
- The scripts and data have been maintained by authors up until July 2020.
- Academic contribution is expected and welcome. Feel free to reach out to authors in case of any query, concern and suggestion.
- Muhammad Saad at University of Central Florida (UCF) and can be approached with email.
- Muhammad Hassan at University of Illinois at Chicago (UIC) and can be approached at this email.